interaction_id stringlengths 9 9 | session_id stringlengths 6 6 | user_id stringclasses 438
values | timestamp_utc stringdate 2025-02-01 02:22:18 2025-04-30 22:23:19 | split stringclasses 3
values | channel stringclasses 5
values | use_case stringclasses 6
values | country_code stringclasses 10
values | region stringclasses 3
values | account_tier stringclasses 3
values | segment stringclasses 3
values | instruction_template stringclasses 6
values | model_provider stringclasses 6
values | model_name stringclasses 8
values | temperature float64 0 1.2 | top_p float64 0.71 1 | max_tokens int64 256 2.05k | prompt_tokens int64 13 1.91k | completion_tokens int64 5 1.12k | total_tokens int64 29 2.75k | latency_ms int64 156 5.62k | cost_usd float64 0 1.05 | response_quality_score float64 0.09 1 | is_failure bool 2
classes | failure_type stringclasses 7
values | hallucination_flag bool 2
classes | toxicity_flag bool 2
classes | safety_block_flag bool 2
classes | formatting_error_flag bool 2
classes | tool_error_flag bool 2
classes | latency_timeout_flag bool 2
classes | tool_calls_count int64 0 3 | tools_used stringclasses 65
values | user_feedback_score float64 -2 2 ⌀ | user_feedback_label stringclasses 4
values | user_reported_issue bool 2
classes | retry_index int64 0 3 | repair_strategy stringclasses 7
values | final_resolution_state stringclasses 3
values | business_impact_tag stringclasses 4
values | request_text stringlengths 63 301 | response_text_snippet stringlengths 180 263 | tokens_per_second float64 27.4 7.66k | date_utc stringdate 2025-02-01 00:00:00 2025-04-30 00:00:00 | hour_of_day_utc int64 0 23 | day_of_week int64 0 6 | is_weekend bool 2
classes | is_peak_hour bool 2
classes | prompt_to_completion_ratio float64 0.02 208 | request_text_template stringclasses 140
values | instruction_text stringclasses 6
values | user_feedback_observed bool 2
classes | prompt_id stringclasses 36
values | response_text_snippet_len_chars int64 180 263 | response_text_snippet_is_truncated bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
INT008801 | S01086 | U00404 | 2025-04-13T22:41:54Z | test | api | coding_assistant | IN | APAC | free | team | internal_policy_qa | openai | gpt-4.1 | 0.231 | 0.841 | 1,024 | 558 | 98 | 656 | 1,765 | 0.2624 | 0.517 | true | formatting_error | false | false | false | true | false | false | 0 | none | null | no_feedback | true | 0 | switch_model_family | escalated_to_human | low | Context for you: I'm getting a JavaScript error: `ValueError`. Can you explain what it means and how to fix it? I'd like a practical, engineering-focused answer. | Let's turn this into a reliable workflow. Signal: formatting_error while working on API retry logic. Quick diagnosis: Use constrained generation (schema-first prompt) and add unit tests for formatting.... | 371.671388 | 2025-04-13 | 22 | 6 | true | false | 5.6939 | I'm getting a JavaScript error: `ValueError`. Can you explain what it means and how to fix it? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | f84eff1764fee544ced395082209fcd5e43e6a0c | 204 | true |
INT008802 | S01025 | U00343 | 2025-04-15T21:44:05Z | train | web_app | content_writing | US | AMER | enterprise | individual | analytics_helper | anthropic | claude-3.5-sonnet | 0.548 | 0.876 | 1,024 | 210 | 388 | 598 | 1,711 | 0.19136 | 1 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | medium | I'm stuck on this and need some guidance: Draft a friendly email announcing a small price increase to our existing customers. Can you walk me through a good solution? | I'll keep this concrete and measurable. Goal: handle evaluation for LLM systems reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating.... | 349.503214 | 2025-04-15 | 21 | 1 | false | false | 0.5412 | Draft a friendly email announcing a small price increase to our existing customers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | 9716fa3b2c6fe39ac0f75ddaad781dfa757a0de9 | 232 | true |
INT008803 | S00589 | U00070 | 2025-03-21T11:16:49Z | train | api | customer_support | US | AMER | free | enterprise_team | creative_writer | meta | llama-3.1-8b | 0.94 | 0.904 | 1,024 | 1,367 | 249 | 1,616 | 716 | 0.09696 | 0.773 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | medium | Context for you: Our customer is getting an error when trying to log in to their account. Can you help me debug this? Can you walk me through a good solution? | I'll keep this concrete and measurable. Goal: handle unexpected charge reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Add telemetry fields... | 2,256.98324 | 2025-03-21 | 11 | 4 | false | true | 5.49 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 7a490bfafd7436aea43a528e9fe4cb6db10b9a6b | 245 | true |
INT008804 | S00045 | U00016 | 2025-02-11T10:45:43Z | train | api | brainstorming | GB | EMEA | free | individual | analytics_helper | meta | llama-3.1-8b | 0.589 | 0.891 | 512 | 1,002 | 407 | 1,409 | 1,552 | 0.08452 | 0.635 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Brainstorm taglines for a productivity app that helps engineers stay focused. | I'll keep this concrete and measurable. Goal: handle cost optimization reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for cost optimization and add... | 907.860825 | 2025-02-11 | 10 | 1 | false | true | 2.4619 | Brainstorm taglines for a productivity app that helps engineers stay focused. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | false | 2df52d847c72d1f6b37061123acb53282123086b | 217 | true |
INT008805 | S01193 | U00265 | 2025-02-17T16:36:49Z | train | api | internal_qa | BR | AMER | enterprise | individual | creative_writer | meta | llama-3.1-8b | 0.644 | 0.954 | 1,024 | 1,023 | 358 | 1,381 | 698 | 0.08284 | 0.335 | true | hallucination | true | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | decompose_task | auto_resolved | medium | Quick question: Check this draft announcement for any confusing or inconsistent statements. Please explain step by step. | We can make this robust with a few changes. Signal: hallucination while working on latency. Quick diagnosis: Refuse when confidence is low; avoid confident guesses. Recommended actions: - Set alert... | 1,978.510029 | 2025-02-17 | 16 | 0 | false | true | 2.8575 | Check this draft announcement for any confusing or inconsistent statements. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | 3ee501c0eea07616af572d79e74d4f247c4b4851 | 200 | true |
INT008806 | S00186 | U00256 | 2025-03-21T05:01:19Z | val | slack | customer_support | CA | AMER | free | team | code_assistant_secure | anthropic | claude-3.5-sonnet | 0.524 | 0.936 | 512 | 212 | 81 | 293 | 643 | 0.09376 | 0.762 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | I'm working on something and need help: A user says they were double charged this month. Help me draft a clear explanation and next steps. | Here are the guardrails I'd put in place. Goal: handle login issues reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Start by writing a small... | 455.676516 | 2025-03-21 | 5 | 4 | false | false | 2.6173 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 1a8e13c1d6c8047debb9a230bf3ae08524ccb884 | 247 | true |
INT008807 | S00590 | U00004 | 2025-04-11T16:25:32Z | val | web_app | brainstorming | US | AMER | free | team | analytics_helper | google | gemini-1.5-pro | 0.51 | 0.876 | 768 | 834 | 290 | 1,124 | 978 | 0.35968 | 0.776 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | medium | I'm stuck on this and need some guidance: Suggest a few content ideas for our next newsletter aimed at early-stage startups. Please explain step by step. | Here's what I'd implement in production. Goal: handle knowledge base search reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. -... | 1,149.284254 | 2025-04-11 | 16 | 4 | false | true | 2.8759 | Suggest a few content ideas for our next newsletter aimed at early-stage startups. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | true | 2df52d847c72d1f6b37061123acb53282123086b | 234 | true |
INT008808 | S01005 | U00162 | 2025-03-14T13:51:06Z | train | slack | customer_support | AE | EMEA | enterprise | enterprise_team | support_strict | mistral | mistral-large | 0.694 | 0.942 | 1,536 | 1,028 | 685 | 1,713 | 672 | 0.20556 | 0.643 | false | none | false | false | false | false | false | false | 3 | code_run|knowledge_lookup|workflow_orchestrator | null | no_feedback | false | 0 | none | auto_resolved | low | This is coming from an internal team in our company. I need a short, friendly reply to a customer who is asking about the status of their refund. I'd like a practical, engineering-focused answer. | We can make this robust with a few changes. Goal: handle data export request reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Start by writing a... | 2,549.107143 | 2025-03-14 | 13 | 4 | false | true | 1.5007 | This is coming from an internal team in our company. I need a short, friendly reply to a customer who is asking about the status of their refund. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | d029d0919f8e306d09ca7001d500c1e6f1c81007 | 250 | true |
INT008809 | S00138 | U00437 | 2025-03-05T16:49:07Z | val | web_app | customer_support | DE | EMEA | free | individual | analytics_helper | anthropic | claude-3.5-sonnet | 0.488 | 0.856 | 1,024 | 974 | 273 | 1,247 | 745 | 0.39904 | 0.98 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | critical | I'm working on something and need help: Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? I'd like a practical, engineering-focused answer. | Here's the clean way to approach it. Goal: handle missing feature reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for missing feature and add tests before optimizing. -... | 1,673.825503 | 2025-03-05 | 16 | 2 | false | true | 3.5678 | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 598eba0ae15195966a9835b68c542f0b34fb88e4 | 237 | true |
INT008810 | S01190 | U00109 | 2025-03-20T16:58:35Z | train | web_app | brainstorming | US | AMER | pro | individual | analytics_helper | openai | gpt-4o-mini | 0.655 | 0.918 | 2,048 | 953 | 115 | 1,068 | 534 | 0.02136 | 0.591 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | abandoned | medium | Quick question: I need naming ideas for an internal tool that tracks experiments and model runs. I'd like a practical, engineering-focused answer. | We can make this robust with a few changes. Goal: handle incident response automation reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for incident response automation and add tests... | 2,000 | 2025-03-20 | 16 | 3 | false | true | 8.287 | I need naming ideas for an internal tool that tracks experiments and model runs. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | true | 2df52d847c72d1f6b37061123acb53282123086b | 249 | true |
INT008811 | S01272 | U00317 | 2025-02-26T11:00:26Z | train | web_app | data_analysis | AU | APAC | enterprise | team | analytics_helper | google | gemini-1.5-pro | 0.348 | 0.882 | 768 | 66 | 356 | 422 | 1,082 | 0.13504 | 1 | false | none | false | false | false | false | false | false | 3 | knowledge_lookup|sql_query|workflow_orchestrator | -1 | negative | false | 0 | none | auto_resolved | critical | Can you help me debug this? Can you suggest a few plots to explore churn risk in this dataset of subscriptions? Can you walk me through a good solution? | I'll keep this concrete and measurable. Goal: handle error rate reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Use schema validation... | 390.018484 | 2025-02-26 | 11 | 2 | false | true | 0.1854 | Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | fbaac57cdb64395ff60fbf79b148e22caaff1431 | 239 | true |
INT008812 | S00527 | U00373 | 2025-03-12T07:40:43Z | train | web_app | data_analysis | DE | EMEA | pro | team | analytics_helper | meta | llama-3.1-70b | 0.266 | 0.836 | 512 | 912 | 104 | 1,016 | 684 | 0.2032 | 0.543 | true | hallucination | true | false | false | false | false | false | 0 | none | null | no_feedback | false | 1 | decompose_task | auto_resolved | low | I'm working on something and need help: Follow-up: the previous reply did not fully solve the issue. Can you suggest a few plots to explore churn risk in this dataset of subscriptions? A short, clear answer would be ideal. | Let's break this down into actionable steps. Signal: hallucination while working on cost per request. Quick diagnosis: Add grounding: retrieve evidence, require citations, and verify key facts.... | 1,485.380117 | 2025-03-12 | 7 | 2 | false | false | 8.7692 | Follow-up: the previous reply did not fully solve the issue. Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | fbaac57cdb64395ff60fbf79b148e22caaff1431 | 196 | true |
INT008813 | S00772 | U00090 | 2025-04-02T13:00:04Z | train | slack | customer_support | AU | APAC | pro | individual | analytics_helper | meta | llama-3.1-8b | 0.567 | 0.913 | 768 | 458 | 275 | 733 | 1,394 | 0.04396 | 0.823 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | I'm stuck on this and need some guidance: A user says they were double charged this month. Help me draft a clear explanation and next steps. Can you walk me through a good solution? | Below is a decision-focused breakdown. Goal: handle missing feature reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for missing feature and add tests before optimizing. - Prefer... | 525.824964 | 2025-04-02 | 13 | 2 | false | true | 1.6655 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 598eba0ae15195966a9835b68c542f0b34fb88e4 | 246 | true |
INT008814 | S00373 | U00291 | 2025-02-11T00:40:23Z | train | web_app | content_writing | IN | APAC | pro | individual | code_assistant_secure | anthropic | claude-3.5-sonnet | 0.66 | 0.83 | 1,536 | 489 | 533 | 1,022 | 777 | 0.32704 | 1 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | auto_resolved | low | Quick question: Rewrite this paragraph in a clearer and more professional tone. | I'll keep this concrete and measurable. Goal: handle prompt injection reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Use schema validation... | 1,315.315315 | 2025-02-11 | 0 | 1 | false | false | 0.9174 | Rewrite this paragraph in a clearer and more professional tone. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | 20f1dad1d39f7b8df6faaa166fe78024f8d03fd6 | 245 | true |
INT008815 | S00417 | U00133 | 2025-03-28T02:25:49Z | train | web_app | brainstorming | FR | EMEA | pro | team | creative_writer | local | custom-local-8b | 0.575 | 0.837 | 768 | 344 | 768 | 1,112 | 741 | 0.00444 | 0.405 | true | hallucination | true | false | false | false | false | false | 0 | none | -1 | negative | false | 0 | decompose_task | auto_resolved | critical | Can you help me debug this? Give me ten creative ideas for a YouTube series about learning data science from scratch. | Here's what I'd implement in production. Signal: hallucination while working on LLMOps dashboards. Quick diagnosis: Use a 'claim -> evidence -> answer' structure and log missing evidence.... | 1,500.674764 | 2025-03-28 | 2 | 4 | false | false | 0.4479 | Give me ten creative ideas for a YouTube series about learning data science from scratch. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | true | 6f52c3b4cdce68aa614a61f324617ae8d0e91690 | 190 | true |
INT008816 | S01040 | U00193 | 2025-03-12T10:54:05Z | val | web_app | content_writing | IN | APAC | free | individual | support_strict | google | gemini-1.5-pro | 0.67 | 0.864 | 256 | 215 | 256 | 471 | 803 | 0.15072 | 0.778 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | auto_resolved | high | I'm stuck on this and need some guidance: Rewrite this paragraph in a clearer and more professional tone. A short, clear answer would be ideal. | Here's the clean way to approach it. Goal: handle tool calling reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for tool calling and add tests before optimizing. - Add telemetry... | 586.550436 | 2025-03-12 | 10 | 2 | false | true | 0.8398 | Rewrite this paragraph in a clearer and more professional tone. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | 75b7366387b6cf4f5631ec9dd298a359e5bb9c1d | 245 | true |
INT008817 | S00412 | U00386 | 2025-03-22T09:36:43Z | test | web_app | internal_qa | IN | APAC | free | individual | internal_policy_qa | openai | gpt-4.1 | 0.732 | 0.993 | 1,024 | 347 | 59 | 406 | 1,511 | 0.1624 | 0.634 | true | hallucination | true | false | false | false | false | false | 0 | none | null | no_feedback | true | 1 | switch_model_family | escalated_to_human | low | I'm working on something and need help: Follow-up: the previous reply did not fully solve the issue. Check this draft announcement for any confusing or inconsistent statements. Can you walk me through a good solution? | Let's make the behavior deterministic and observable. Signal: hallucination while working on provider degradation. Quick diagnosis: Add grounding: retrieve evidence, require citations, and verify key facts.... | 268.696228 | 2025-03-22 | 9 | 5 | true | true | 5.8814 | Follow-up: the previous reply did not fully solve the issue. Check this draft announcement for any confusing or inconsistent statements. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | 8ad879fd7895b38236ec6ff25a956251dc06e51b | 209 | true |
INT008818 | S01020 | U00167 | 2025-04-15T11:26:43Z | train | mobile_app | customer_support | IN | APAC | free | team | general_assistant | google | gemini-1.5-pro | 0.729 | 0.921 | 512 | 280 | 469 | 749 | 852 | 0.23968 | 0.751 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | abandoned | medium | I'm working on something and need help: A user says they were double charged this month. Help me draft a clear explanation and next steps. Can you walk me through a good solution? | Let's make the behavior deterministic and observable. Goal: handle unexpected charge reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from... | 879.107981 | 2025-04-15 | 11 | 1 | false | true | 0.597 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | a52fb80193c7c0769c0836852624f4b025a04d2a | 224 | true |
INT008819 | S00506 | U00171 | 2025-03-04T10:23:23Z | test | mobile_app | data_analysis | GB | EMEA | free | team | internal_policy_qa | mistral | mistral-large | 0.28 | 0.843 | 512 | 338 | 512 | 850 | 735 | 0.102 | 0.577 | true | tool_error | false | false | false | false | true | false | 2 | knowledge_lookup|sql_query | -2 | negative | false | 1 | add_constraints_to_prompt | escalated_to_human | low | Context for you: Follow-up: the previous reply did not fully solve the issue. Help me compute the conversion rate per marketing channel and highlight the top 3 performers. Can you walk me through a good solution? | Let's make the behavior deterministic and observable. Signal: tool_error while working on token usage. Quick diagnosis: Validate tool arguments before calling the tool; reject invalid calls early.... | 1,156.462585 | 2025-03-04 | 10 | 1 | false | true | 0.6602 | Follow-up: the previous reply did not fully solve the issue. Help me compute the conversion rate per marketing channel and highlight the top 3 performers. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 62646e5a7026d4f68ab1867a730451326aa3b664 | 199 | true |
INT008820 | S00479 | U00158 | 2025-03-12T00:50:58Z | train | internal_tool | coding_assistant | IN | APAC | free | individual | support_strict | openai | gpt-4.1 | 0.266 | 0.856 | 1,024 | 775 | 591 | 1,366 | 949 | 0.5464 | 0.645 | true | tool_error | false | false | false | false | true | false | 1 | tool_invocation_failed | null | no_feedback | false | 0 | regenerate_with_lower_temp | abandoned | low | I'm working on something and need help: I'm getting a Java error: `IndexError`. Can you explain what it means and how to fix it? Please explain step by step. | Here's the clean way to approach it. Let's stabilize it. Signal: tool_error while working on API retry logic. Quick diagnosis: Validate tool arguments before calling the tool; reject invalid calls early. Recommended actions: - Slice... | 1,439.409905 | 2025-03-12 | 0 | 2 | false | false | 1.3113 | I'm getting a Java error: `IndexError`. Can you explain what it means and how to fix it? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | da34aac46ee8d6d44012043be92e33884c6b4920 | 235 | true |
INT008821 | S00295 | U00345 | 2025-02-28T00:28:57Z | train | internal_tool | customer_support | EG | EMEA | free | individual | internal_policy_qa | anthropic | claude-3.5-sonnet | 0.453 | 0.901 | 1,024 | 825 | 111 | 936 | 1,340 | 0.29952 | 0.63 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | medium | I'm stuck on this and need some guidance: Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? Please explain step by step. | Here's a clear, step-by-step answer. Goal: handle slow responses reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Use schema... | 698.507463 | 2025-02-28 | 0 | 4 | false | false | 7.4324 | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 03bf57d7adc83026d525c8192379aa203b71e1eb | 229 | true |
INT008822 | S00168 | U00447 | 2025-03-19T17:03:00Z | train | slack | customer_support | AE | EMEA | free | team | creative_writer | local | custom-local-8b | 0.653 | 0.831 | 768 | 405 | 768 | 1,173 | 1,303 | 0.00468 | 0.919 | false | none | false | false | false | false | false | false | 3 | http_request|sql_query|vector_search | 1 | positive | false | 0 | none | auto_resolved | high | Context for you: A user says they were double charged this month. Help me draft a clear explanation and next steps. Can you walk me through a good solution? | I'll keep this concrete and measurable. Goal: handle slow responses reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for slow responses and add tests... | 900.230238 | 2025-03-19 | 17 | 2 | false | true | 0.5273 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 7a490bfafd7436aea43a528e9fe4cb6db10b9a6b | 217 | true |
INT008823 | S00359 | U00159 | 2025-02-18T09:41:55Z | train | api | coding_assistant | FR | EMEA | enterprise | team | analytics_helper | local | custom-local-8b | 0.238 | 0.902 | 1,536 | 821 | 365 | 1,186 | 1,493 | 0.00476 | 0.876 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | auto_resolved | critical | I'm stuck on this and need some guidance: I'm getting a C# error: `NullPointerException`. Can you explain what it means and how to fix it? Please explain step by step. | Below is a decision-focused breakdown. Goal: handle JSON schema validation reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Add telemetry... | 794.373744 | 2025-02-18 | 9 | 1 | false | true | 2.2493 | I'm getting a C# error: `NullPointerException`. Can you explain what it means and how to fix it? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | 54f94ac4f6df7ccc4808605ae4fb839b7a1fbfb2 | 242 | true |
INT008824 | S00509 | U00331 | 2025-03-26T16:19:53Z | train | web_app | customer_support | FR | EMEA | free | individual | code_assistant_secure | mistral | mistral-large | 0.447 | 0.971 | 1,536 | 396 | 135 | 531 | 1,585 | 0.06372 | 0.75 | false | none | false | false | false | false | false | false | 3 | http_request|knowledge_lookup|sql_query | 2 | positive | false | 0 | none | escalated_to_human | low | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? Please explain step by step. | This is a common pattern; the fix is straightforward. Goal: handle slow responses reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer... | 335.015773 | 2025-03-26 | 16 | 2 | false | true | 2.9333 | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 1a8e13c1d6c8047debb9a230bf3ae08524ccb884 | 243 | true |
INT008825 | S00726 | U00065 | 2025-04-19T10:36:03Z | test | web_app | customer_support | GB | EMEA | enterprise | individual | code_assistant_secure | meta | llama-3.1-70b | 0.43 | 0.894 | 1,536 | 663 | 706 | 1,369 | 833 | 0.2738 | 0.671 | false | none | false | false | false | false | false | false | 1 | http_request | 0 | neutral | false | 0 | none | auto_resolved | medium | Can you help me debug this? A user says they were double charged this month. Help me draft a clear explanation and next steps. Please explain step by step. | Below is a practical plan you can execute. Goal: handle billing dispute reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer... | 1,643.457383 | 2025-04-19 | 10 | 5 | true | true | 0.9391 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 1a8e13c1d6c8047debb9a230bf3ae08524ccb884 | 233 | true |
INT008826 | S00329 | U00004 | 2025-02-20T22:51:42Z | train | mobile_app | internal_qa | US | AMER | free | team | analytics_helper | openai | gpt-4o-mini | 0.949 | 0.943 | 1,024 | 336 | 530 | 866 | 1,564 | 0.01732 | 0.98 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | abandoned | low | I'm working on something and need help: Summarise this internal security policy into three bullet points for non-technical employees. A short, clear answer would be ideal. | I'll keep this concrete and measurable. Goal: handle token spikes reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for token spikes and add tests before... | 553.70844 | 2025-02-20 | 22 | 3 | false | false | 0.634 | Summarise this internal security policy into three bullet points for non-technical employees. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | d33eca392020f2e6e14ec7fe3091a25b46d6bae1 | 220 | true |
INT008827 | S01364 | U00093 | 2025-02-02T00:40:46Z | train | web_app | customer_support | US | AMER | pro | individual | support_strict | meta | llama-3.1-70b | 0.593 | 0.91 | 768 | 675 | 493 | 1,168 | 2,142 | 0.2336 | 0.623 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | abandoned | low | Quick question: I need a short, friendly reply to a customer who is asking about the status of their refund. Please explain step by step. | Let's turn this into a reliable workflow. Goal: handle missing feature reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from... | 545.284781 | 2025-02-02 | 0 | 6 | true | false | 1.3692 | I need a short, friendly reply to a customer who is asking about the status of their refund. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | d029d0919f8e306d09ca7001d500c1e6f1c81007 | 210 | true |
INT008828 | S00570 | U00409 | 2025-04-18T14:46:46Z | train | slack | customer_support | AE | EMEA | free | team | analytics_helper | openai | gpt-4.1 | 0.869 | 0.966 | 2,048 | 865 | 344 | 1,209 | 1,341 | 0.4836 | 0.549 | true | tool_error | false | false | false | false | true | false | 1 | tool_invocation_failed | -1 | negative | true | 1 | regenerate_with_lower_temp | abandoned | medium | I'm working on something and need help: Follow-up: the previous reply did not fully solve the issue. Our customer is getting an error when trying to log in to their account. Can you help me debug this? | Let's break this down into actionable steps. Signal: tool_error while working on account verification. Quick diagnosis: Validate tool arguments before calling the tool; reject invalid calls early. Recommended... | 901.565996 | 2025-04-18 | 14 | 4 | false | true | 2.5145 | Follow-up: the previous reply did not fully solve the issue. Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 598eba0ae15195966a9835b68c542f0b34fb88e4 | 211 | true |
INT008829 | S01326 | U00389 | 2025-04-01T19:40:48Z | train | web_app | coding_assistant | EG | EMEA | free | individual | general_assistant | meta | llama-3.1-8b | 0.334 | 0.88 | 1,536 | 913 | 448 | 1,361 | 1,408 | 0.08164 | 0.872 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 1 | regenerate_with_lower_temp | auto_resolved | low | Quick question: Follow-up: the previous reply did not fully solve the issue. I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. I'd like a practical, engineering-focused answer. | Here's a clear, step-by-step answer. Goal: handle async I/O pitfalls reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Start by... | 966.619318 | 2025-04-01 | 19 | 1 | false | false | 2.0379 | Follow-up: the previous reply did not fully solve the issue. I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 3ddc6acb5986d68a027a4c760a5adfbe894b8deb | 236 | true |
INT008830 | S00077 | U00396 | 2025-04-11T18:27:01Z | test | web_app | data_analysis | US | AMER | enterprise | enterprise_team | creative_writer | anthropic | claude-3.5-sonnet | 0.385 | 0.888 | 1,024 | 456 | 25 | 481 | 2,646 | 0.15392 | 0.813 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | abandoned | critical | I'm stuck on this and need some guidance: I want a short summary of key metrics for this dataset: total users, active users, and average revenue per user. I'd like a practical, engineering-focused answer. | We can make this robust with a few changes. Goal: handle cost per request reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Start by writing... | 181.783825 | 2025-04-11 | 18 | 4 | false | true | 18.24 | I want a short summary of key metrics for this dataset: total users, active users, and average revenue per user. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | 814335a175319858ca540d38bfcac020ec643c3f | 249 | true |
INT008831 | S00873 | U00288 | 2025-04-01T19:08:48Z | train | api | internal_qa | FR | EMEA | pro | enterprise_team | internal_policy_qa | meta | llama-3.1-8b | 0.1 | 0.863 | 1,024 | 468 | 367 | 835 | 1,041 | 0.05008 | 0.453 | true | formatting_error | false | false | false | true | false | false | 0 | none | null | no_feedback | false | 0 | manual_post_edit | abandoned | medium | Context for you: Turn this long internal announcement into a short FAQ-style document. I'd like a practical, engineering-focused answer. | Below is a practical plan you can execute. Signal: formatting_error while working on tool failures. Quick diagnosis: Use constrained generation (schema-first prompt) and add unit tests for... | 802.113353 | 2025-04-01 | 19 | 1 | false | false | 1.2752 | Turn this long internal announcement into a short FAQ-style document. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | 8ad879fd7895b38236ec6ff25a956251dc06e51b | 191 | true |
INT008832 | S01545 | U00154 | 2025-04-08T17:31:18Z | train | web_app | data_analysis | GB | EMEA | free | individual | creative_writer | local | custom-local-8b | 0.209 | 0.961 | 1,536 | 735 | 82 | 817 | 536 | 0.00328 | 0.98 | false | none | false | false | false | false | false | false | 1 | workflow_orchestrator | 1 | positive | false | 0 | none | abandoned | low | I'm working on something and need help: I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. Can you walk me through a good solution? | Let's break this down into actionable steps. Goal: handle cost per request reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer... | 1,524.253731 | 2025-04-08 | 17 | 1 | false | true | 8.9634 | I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 814335a175319858ca540d38bfcac020ec643c3f | 236 | true |
INT008833 | S00818 | U00263 | 2025-03-06T00:58:40Z | train | api | customer_support | AU | APAC | free | team | internal_policy_qa | local | custom-local-8b | 0.698 | 0.926 | 1,024 | 1,111 | 515 | 1,626 | 1,494 | 0.00652 | 0.744 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | I'm working on something and need help: Our customer is getting an error when trying to log in to their account. Can you help me debug this? A short, clear answer would be ideal. | Let's break this down into actionable steps. Goal: handle account verification reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure... | 1,088.353414 | 2025-03-06 | 0 | 3 | false | false | 2.1573 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 03bf57d7adc83026d525c8192379aa203b71e1eb | 222 | true |
INT008834 | S01524 | U00158 | 2025-04-05T04:03:16Z | train | web_app | content_writing | IN | APAC | free | individual | internal_policy_qa | openai | gpt-4.1 | 0.734 | 0.964 | 256 | 161 | 88 | 249 | 1,289 | 0.0996 | 0.537 | true | safety_block | false | false | true | false | false | false | 0 | none | -2 | negative | false | 2 | regenerate_with_lower_temp | escalated_to_human | high | I'm working on something and need help: Follow-up: the previous reply did not fully solve the issue. Turn these bullet points into a short, engaging blog introduction about learning machine learning. I'd like a practical, engineering-focused answer. | Let's break this down into actionable steps. Signal: safety_block while working on LLM reliability. Quick diagnosis: Refuse safely and offer a legitimate alternative or a high-level explanation.... | 193.173002 | 2025-04-05 | 4 | 5 | true | false | 1.8295 | Follow-up: the previous reply did not fully solve the issue. Turn these bullet points into a short, engaging blog introduction about learning machine learning. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | db47a2ba00ac8f8120dcd2fd47d3648a8f594c66 | 197 | true |
INT008835 | S00198 | U00253 | 2025-04-29T01:43:02Z | train | internal_tool | brainstorming | IN | APAC | free | individual | internal_policy_qa | meta | llama-3.1-8b | 0.857 | 0.953 | 2,048 | 1,358 | 683 | 2,041 | 944 | 0.12244 | 0.474 | true | hallucination | true | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | decompose_task | abandoned | critical | Can you help me debug this? Brainstorm taglines for a productivity app that helps engineers stay focused. I'd like a practical, engineering-focused answer. | Let's break this down into actionable steps. Signal: hallucination while working on knowledge base search. Quick diagnosis: Refuse when confidence is low; avoid confident guesses. Recommended actions: - Add unit... | 2,162.076271 | 2025-04-29 | 1 | 1 | false | false | 1.9883 | Brainstorm taglines for a productivity app that helps engineers stay focused. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | false | 64f5ffde54eb84c78cab2d2147174c299cd15237 | 214 | true |
INT008836 | S00589 | U00070 | 2025-03-21T11:19:25Z | train | api | customer_support | US | AMER | free | enterprise_team | general_assistant | meta | llama-3.1-8b | 0.237 | 0.865 | 512 | 391 | 337 | 728 | 667 | 0.04368 | 0.732 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Quick question: I need a short, friendly reply to a customer who is asking about the status of their refund. Please explain step by step. | Here's what I'd implement in production. Goal: handle unexpected charge reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for unexpected charge and add tests before optimizing. -... | 1,091.454273 | 2025-03-21 | 11 | 4 | false | true | 1.1602 | I need a short, friendly reply to a customer who is asking about the status of their refund. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | a52fb80193c7c0769c0836852624f4b025a04d2a | 245 | true |
INT008837 | S00782 | U00131 | 2025-04-21T16:01:44Z | train | api | data_analysis | CA | AMER | pro | individual | creative_writer | openai | gpt-4o-mini | 0.395 | 0.957 | 2,048 | 1,270 | 211 | 1,481 | 694 | 0.02964 | 0.694 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Context for you: I want a short summary of key metrics for this dataset: total users, active users, and average revenue per user. Can you walk me through a good solution? | We can make this robust with a few changes. Goal: handle latency p95 reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. -... | 2,134.005764 | 2025-04-21 | 16 | 0 | false | true | 6.019 | I want a short summary of key metrics for this dataset: total users, active users, and average revenue per user. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | 814335a175319858ca540d38bfcac020ec643c3f | 222 | true |
INT008838 | S00335 | U00258 | 2025-03-10T05:40:22Z | train | web_app | content_writing | GB | EMEA | free | team | creative_writer | local | custom-local-8b | 0.821 | 0.957 | 1,024 | 361 | 301 | 662 | 753 | 0.00264 | 0.677 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | abandoned | high | Can you help me debug this? Write a concise product description for a new wireless mouse aimed at software developers. A short, clear answer would be ideal. | Here's what I'd implement in production. Goal: handle observability metrics reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Start by... | 879.150066 | 2025-03-10 | 5 | 0 | false | false | 1.1993 | Write a concise product description for a new wireless mouse aimed at software developers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | 469eb15f8b08d9b4130e14a6c3888f14e335a854 | 243 | true |
INT008839 | S01462 | U00096 | 2025-03-24T22:55:21Z | train | slack | brainstorming | CA | AMER | pro | individual | general_assistant | openai | gpt-4o-mini | 0.582 | 0.824 | 1,024 | 546 | 751 | 1,297 | 942 | 0.02592 | 0.808 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | abandoned | high | Brainstorm taglines for a productivity app that helps engineers stay focused. A short, clear answer would be ideal. | Below is a practical plan you can execute. Goal: handle cost optimization reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for cost optimization and add... | 1,376.857749 | 2025-03-24 | 22 | 0 | false | false | 0.727 | Brainstorm taglines for a productivity app that helps engineers stay focused. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | false | 62cac6cf36e4ae3f199823b7f2ca46aa5e4a0aa7 | 220 | true |
INT008840 | S00503 | U00063 | 2025-03-25T21:54:00Z | test | api | content_writing | BR | AMER | free | individual | analytics_helper | google | gemini-1.5-pro | 0.865 | 0.861 | 2,048 | 731 | 466 | 1,197 | 1,279 | 0.38304 | 0.745 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | abandoned | high | Quick question: Draft a friendly email announcing a small price increase to our existing customers. Please explain step by step. | Here's the clean way to approach it. Goal: handle A/B testing reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for A/B testing and add tests before optimizing. - Use schema... | 935.887412 | 2025-03-25 | 21 | 1 | false | false | 1.5687 | Draft a friendly email announcing a small price increase to our existing customers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | 9716fa3b2c6fe39ac0f75ddaad781dfa757a0de9 | 240 | true |
INT008841 | S01486 | U00431 | 2025-03-06T22:15:59Z | val | web_app | data_analysis | US | AMER | free | enterprise_team | analytics_helper | openai | gpt-4o-mini | 0.231 | 0.965 | 2,048 | 1,063 | 516 | 1,579 | 1,523 | 0.03156 | 0.443 | true | hallucination | true | false | false | false | false | false | 0 | none | -2 | negative | true | 0 | regenerate_with_lower_temp | escalated_to_human | medium | Context for you: Can you suggest a few plots to explore churn risk in this dataset of subscriptions? I'd like a practical, engineering-focused answer. | Let's break this down into actionable steps. Signal: hallucination while working on safety flag rate. Quick diagnosis: Add grounding: retrieve evidence, require citations, and verify key facts.... | 1,036.769534 | 2025-03-06 | 22 | 3 | false | false | 2.0601 | Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | fbaac57cdb64395ff60fbf79b148e22caaff1431 | 196 | true |
INT008842 | S00491 | U00021 | 2025-02-05T05:14:06Z | train | web_app | customer_support | AU | APAC | pro | team | analytics_helper | meta | llama-3.1-70b | 0.593 | 0.814 | 768 | 480 | 768 | 1,248 | 900 | 0.2496 | 0.536 | true | hallucination | true | false | false | false | false | false | 0 | none | -2 | negative | true | 0 | add_constraints_to_prompt | escalated_to_human | low | I'm stuck on this and need some guidance: I need a short, friendly reply to a customer who is asking about the status of their refund. | This is a common pattern; the fix is straightforward. Signal: hallucination while working on billing dispute. Quick diagnosis: Use a 'claim -> evidence -> answer' structure and log missing evidence. Recommended actions:... | 1,386.666667 | 2025-02-05 | 5 | 2 | false | false | 0.625 | I need a short, friendly reply to a customer who is asking about the status of their refund. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 598eba0ae15195966a9835b68c542f0b34fb88e4 | 222 | true |
INT008843 | S00235 | U00352 | 2025-04-09T09:14:10Z | val | internal_tool | customer_support | FR | EMEA | enterprise | individual | internal_policy_qa | meta | llama-3.1-8b | 0.492 | 0.938 | 1,024 | 1,676 | 627 | 2,303 | 477 | 0.1382 | 0.65 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Quick question: This is coming from an internal team in our company. A user says they were double charged this month. Help me draft a clear explanation and next steps. Please explain step by step. | Let's turn this into a reliable workflow. Goal: handle data export request reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Add telemetry fields... | 4,828.092243 | 2025-04-09 | 9 | 2 | false | true | 2.673 | This is coming from an internal team in our company. A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 03bf57d7adc83026d525c8192379aa203b71e1eb | 249 | true |
INT008844 | S01344 | U00062 | 2025-02-20T16:30:11Z | train | web_app | coding_assistant | US | AMER | free | individual | general_assistant | google | gemini-1.5-pro | 0.385 | 0.841 | 1,536 | 216 | 295 | 511 | 863 | 0.16352 | 0.695 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | medium | I'm stuck on this and need some guidance: Can you refactor this function to be more readable and efficient while keeping the same behaviour? | Here are the guardrails I'd put in place. Goal: handle API retry logic reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for API retry logic and add tests... | 592.12051 | 2025-02-20 | 16 | 3 | false | true | 0.7322 | Can you refactor this function to be more readable and efficient while keeping the same behaviour? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 3ddc6acb5986d68a027a4c760a5adfbe894b8deb | 221 | true |
INT008845 | S00584 | U00161 | 2025-04-10T01:18:40Z | train | web_app | content_writing | FR | EMEA | free | individual | analytics_helper | meta | llama-3.1-70b | 1.058 | 0.979 | 2,048 | 268 | 419 | 687 | 1,848 | 0.1374 | 0.282 | true | hallucination | true | false | false | false | false | false | 0 | none | null | no_feedback | true | 0 | switch_model_family | escalated_to_human | high | Quick question: Turn these bullet points into a short, engaging blog introduction about learning machine learning. A short, clear answer would be ideal. | Here are the guardrails I'd put in place. Signal: hallucination while working on evaluation for LLM systems. Quick diagnosis: Add grounding: retrieve evidence, require citations, and verify key facts.... | 371.753247 | 2025-04-10 | 1 | 3 | false | false | 0.6396 | Turn these bullet points into a short, engaging blog introduction about learning machine learning. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | 9716fa3b2c6fe39ac0f75ddaad781dfa757a0de9 | 203 | true |
INT008846 | S00080 | U00396 | 2025-04-21T22:10:37Z | train | web_app | coding_assistant | US | AMER | enterprise | enterprise_team | creative_writer | mistral | mistral-large | 0.368 | 0.894 | 512 | 309 | 512 | 821 | 663 | 0.09852 | 0.676 | true | safety_block | false | false | true | false | false | false | 3 | code_run|http_request|vector_search | 0 | neutral | false | 0 | switch_model_family | abandoned | medium | Context for you: Write a small example that shows how to handle exceptions properly in Go. | Below is a practical plan you can execute. Signal: safety_block while working on caching strategy. Quick diagnosis: Route to a safe-help template and suggest benign resources or best practices. Recommended... | 1,238.310709 | 2025-04-21 | 22 | 0 | false | false | 0.6035 | Write a small example that shows how to handle exceptions properly in Go. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | 83c939e9b891e505ac06548e158556e919317a64 | 208 | true |
INT008847 | S00432 | U00231 | 2025-02-12T09:34:07Z | train | mobile_app | content_writing | EG | EMEA | free | team | internal_policy_qa | anthropic | claude-3.5-sonnet | 0.839 | 0.853 | 1,024 | 224 | 170 | 394 | 941 | 0.12608 | 0.618 | false | none | false | false | false | false | false | false | 0 | none | 0 | neutral | false | 0 | none | auto_resolved | high | I'm working on something and need help: Write a concise product description for a new wireless mouse aimed at software developers. | Let's make the behavior deterministic and observable. Goal: handle evaluation for LLM systems reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions... | 418.703507 | 2025-02-12 | 9 | 2 | false | true | 1.3176 | Write a concise product description for a new wireless mouse aimed at software developers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | db47a2ba00ac8f8120dcd2fd47d3648a8f594c66 | 241 | true |
INT008848 | S00180 | U00041 | 2025-02-11T18:02:40Z | train | web_app | brainstorming | AE | EMEA | free | team | general_assistant | google | gemini-1.5-pro | 0.356 | 0.909 | 256 | 984 | 72 | 1,056 | 852 | 0.33792 | 0.675 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | auto_resolved | low | Can you help me debug this? I need naming ideas for an internal tool that tracks experiments and model runs. Please explain step by step. | Here's the clean way to approach it. Goal: handle developer productivity reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer simple... | 1,239.43662 | 2025-02-11 | 18 | 1 | false | true | 13.6667 | I need naming ideas for an internal tool that tracks experiments and model runs. | You are a creative assistant helping users explore ideas and options. Suggest multiple alternatives, structure them clearly, and stay practical. | true | 62cac6cf36e4ae3f199823b7f2ca46aa5e4a0aa7 | 241 | true |
INT008849 | S01500 | U00057 | 2025-03-11T17:21:52Z | train | web_app | coding_assistant | GB | EMEA | free | enterprise_team | code_assistant_secure | local | custom-local-8b | 0.187 | 0.848 | 512 | 896 | 242 | 1,138 | 784 | 0.00456 | 0.824 | false | none | false | false | false | false | false | false | 1 | sql_query | null | no_feedback | false | 0 | none | abandoned | high | I'm stuck on this and need some guidance: I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. Can you walk me through a good solution? | Let's break this down into actionable steps. Goal: handle unit test mocking reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Start by writing a... | 1,451.530612 | 2025-03-11 | 17 | 1 | false | true | 3.7025 | I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 52eadb52b3319d2dab569ce8fb653a0bee56a90f | 253 | true |
INT008850 | S01364 | U00093 | 2025-02-02T00:46:15Z | train | web_app | customer_support | US | AMER | pro | individual | code_assistant_secure | meta | llama-3.1-8b | 0.438 | 0.854 | 768 | 332 | 460 | 792 | 906 | 0.04752 | 0.993 | false | none | false | false | false | false | false | false | 3 | code_run|email_send|knowledge_lookup | null | no_feedback | false | 0 | none | abandoned | low | Can you help me debug this? A user says they were double charged this month. Help me draft a clear explanation and next steps. Can you walk me through a good solution? | I'll keep this concrete and measurable. Goal: handle account verification reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Add telemetry... | 874.172185 | 2025-02-02 | 0 | 6 | true | false | 0.7217 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 1a8e13c1d6c8047debb9a230bf3ae08524ccb884 | 241 | true |
INT008851 | S01525 | U00284 | 2025-04-24T08:31:40Z | train | slack | customer_support | AE | EMEA | enterprise | individual | code_assistant_secure | openai | gpt-4o-mini | 0.921 | 0.967 | 512 | 1,095 | 512 | 1,607 | 1,475 | 0.03212 | 0.833 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | auto_resolved | low | Quick question: This is coming from an internal team in our company. Our customer is getting an error when trying to log in to their account. Can you help me debug this? A short, clear answer would be ideal. | Let's turn this into a reliable workflow. Goal: handle billing dispute reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for billing dispute and add tests before... | 1,089.491525 | 2025-04-24 | 8 | 3 | false | false | 2.1387 | This is coming from an internal team in our company. Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 1a8e13c1d6c8047debb9a230bf3ae08524ccb884 | 228 | true |
INT008852 | S01461 | U00435 | 2025-03-06T22:26:52Z | val | slack | coding_assistant | BR | AMER | pro | team | code_assistant_secure | meta | llama-3.1-70b | 0.311 | 0.934 | 1,024 | 56 | 515 | 571 | 1,440 | 0.1142 | 0.603 | true | safety_block | false | false | true | false | false | false | 1 | workflow_orchestrator | null | no_feedback | false | 0 | regenerate_with_lower_temp | abandoned | low | Write a small example that shows how to handle exceptions properly in JavaScript. A short, clear answer would be ideal. | I'll keep this concrete and measurable. Signal: safety_block while working on async I/O pitfalls. Quick diagnosis: Route to a safe-help template and suggest benign resources or best practices. Recommended... | 396.527778 | 2025-03-06 | 22 | 3 | false | false | 0.1087 | Write a small example that shows how to handle exceptions properly in JavaScript. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 52eadb52b3319d2dab569ce8fb653a0bee56a90f | 207 | true |
INT008853 | S00676 | U00352 | 2025-04-25T06:05:34Z | train | web_app | data_analysis | FR | EMEA | enterprise | individual | creative_writer | openai | gpt-4o-mini | 0.416 | 0.889 | 512 | 1,246 | 174 | 1,420 | 660 | 0.0284 | 0.707 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | abandoned | low | I'm stuck on this and need some guidance: Can you suggest a few plots to explore churn risk in this dataset of subscriptions? I'd like a practical, engineering-focused answer. | I'll keep this concrete and measurable. Goal: handle token usage reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Start by writing a small... | 2,151.515152 | 2025-04-25 | 6 | 4 | false | false | 7.1609 | Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | 814335a175319858ca540d38bfcac020ec643c3f | 248 | true |
INT008854 | S01052 | U00012 | 2025-03-29T10:17:50Z | train | internal_tool | customer_support | AU | APAC | free | team | creative_writer | meta | llama-3.1-8b | 0.1 | 0.912 | 512 | 32 | 42 | 74 | 1,115 | 0.00444 | 0.711 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | low | I'm stuck on this and need some guidance: I need a short, friendly reply to a customer who is asking about the status of their refund. I'd like a practical, engineering-focused answer. | We can make this robust with a few changes. Goal: handle account verification reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for account verification and add tests before... | 66.367713 | 2025-03-29 | 10 | 5 | true | true | 0.7619 | I need a short, friendly reply to a customer who is asking about the status of their refund. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 7a490bfafd7436aea43a528e9fe4cb6db10b9a6b | 240 | true |
INT008855 | S01256 | U00186 | 2025-03-11T19:00:45Z | train | mobile_app | customer_support | IN | APAC | free | enterprise_team | internal_policy_qa | meta | llama-3.1-70b | 0.506 | 0.918 | 1,024 | 726 | 494 | 1,220 | 1,104 | 0.244 | 0.551 | true | formatting_error | false | false | false | true | false | false | 0 | none | -1 | negative | true | 1 | add_constraints_to_prompt | escalated_to_human | medium | Follow-up: the previous reply did not fully solve the issue. A user says they were double charged this month. Help me draft a clear explanation and next steps. I'd like a practical, engineering-focused answer. | Let's break this down into actionable steps. Let's stabilize it. Signal: formatting_error while working on billing dispute. Quick diagnosis: Use constrained generation (schema-first prompt) and add unit tests for formatting. Recommended... | 1,105.072464 | 2025-03-11 | 19 | 1 | false | false | 1.4696 | Follow-up: the previous reply did not fully solve the issue. A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 03bf57d7adc83026d525c8192379aa203b71e1eb | 239 | true |
INT008856 | S00079 | U00347 | 2025-03-04T00:45:30Z | train | web_app | content_writing | US | AMER | free | team | code_assistant_secure | meta | llama-3.1-70b | 0.669 | 0.91 | 512 | 199 | 311 | 510 | 1,171 | 0.102 | 0.475 | true | tool_error | false | false | false | false | true | false | 1 | tool_invocation_failed | null | no_feedback | false | 0 | manual_post_edit | auto_resolved | low | Write a concise product description for a new wireless mouse aimed at software developers. A short, clear answer would be ideal. | We can make this robust with a few changes. Signal: tool_error while working on tool calling. Quick diagnosis: Validate tool arguments before calling the tool; reject invalid calls early. Recommended... | 435.525192 | 2025-03-04 | 0 | 1 | false | false | 0.6399 | Write a concise product description for a new wireless mouse aimed at software developers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | 20f1dad1d39f7b8df6faaa166fe78024f8d03fd6 | 202 | true |
INT008857 | S01574 | U00320 | 2025-02-17T10:18:24Z | train | web_app | customer_support | IN | APAC | free | team | support_strict | meta | llama-3.1-70b | 0.615 | 0.897 | 512 | 656 | 512 | 1,168 | 1,934 | 0.2336 | 0.409 | true | tool_error | false | false | false | false | true | false | 3 | email_send|knowledge_lookup|sql_query | -1 | negative | true | 2 | switch_model_family | escalated_to_human | medium | Can you help me debug this? Follow-up: the previous reply did not fully solve the issue. Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? Please explain step by step. | I'll keep this concrete and measurable. Signal: tool_error while working on missing feature. Quick diagnosis: Implement a fallback path when the tool is degraded. Recommended actions: - Log... | 603.929679 | 2025-02-17 | 10 | 0 | false | true | 1.2812 | Follow-up: the previous reply did not fully solve the issue. Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | d029d0919f8e306d09ca7001d500c1e6f1c81007 | 192 | true |
INT008858 | S00902 | U00262 | 2025-04-04T01:36:42Z | test | mobile_app | customer_support | IN | APAC | free | enterprise_team | internal_policy_qa | openai | gpt-4.1 | 0.543 | 0.899 | 2,048 | 583 | 548 | 1,131 | 1,405 | 0.4524 | 0.791 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Context for you: Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? I'd like a practical, engineering-focused answer. | Here's the clean way to approach it. Goal: handle slow responses reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices.... | 804.982206 | 2025-04-04 | 1 | 4 | false | false | 1.0639 | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 03bf57d7adc83026d525c8192379aa203b71e1eb | 216 | true |
INT008859 | S01338 | U00254 | 2025-02-11T19:23:19Z | test | mobile_app | customer_support | DE | EMEA | pro | team | support_strict | openai | gpt-4.1 | 0.497 | 0.922 | 512 | 166 | 512 | 678 | 2,567 | 0.2712 | 0.794 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | medium | Can you help me debug this? Our customer is getting an error when trying to log in to their account. Can you help me debug this? Can you walk me through a good solution? | Here's a clear, step-by-step answer. Goal: handle data export request reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Add telemetry... | 264.121543 | 2025-02-11 | 19 | 1 | false | false | 0.3242 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | d029d0919f8e306d09ca7001d500c1e6f1c81007 | 238 | true |
INT008860 | S00780 | U00327 | 2025-04-13T11:05:45Z | train | internal_tool | customer_support | DE | EMEA | free | individual | internal_policy_qa | openai | gpt-4o-mini | 0.759 | 0.98 | 512 | 957 | 228 | 1,185 | 1,127 | 0.02372 | 0.645 | false | none | false | false | false | false | false | false | 1 | http_request | null | no_feedback | false | 0 | none | auto_resolved | low | Can you help me debug this? Our customer is getting an error when trying to log in to their account. Can you help me debug this? | Below is a practical plan you can execute. Goal: handle login issues reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Start by... | 1,051.464064 | 2025-04-13 | 11 | 6 | true | true | 4.1974 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 03bf57d7adc83026d525c8192379aa203b71e1eb | 231 | true |
INT008861 | S00354 | U00233 | 2025-02-21T10:57:48Z | val | web_app | data_analysis | EG | EMEA | enterprise | individual | code_assistant_secure | anthropic | claude-3.5-sonnet | 0.265 | 0.746 | 1,536 | 949 | 294 | 1,243 | 1,700 | 0.39776 | 0.759 | false | none | false | false | false | false | false | false | 0 | none | -1 | negative | false | 0 | none | auto_resolved | low | Can you help me debug this? I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | Below is a practical plan you can execute. Goal: handle safety flag rate reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer simple baselines... | 731.176471 | 2025-02-21 | 10 | 4 | false | true | 3.2279 | I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 4e275efe35fdc513983e7d3f3ad32b50cc391969 | 251 | true |
INT008862 | S00599 | U00013 | 2025-03-07T22:27:41Z | train | slack | customer_support | BR | AMER | pro | enterprise_team | general_assistant | anthropic | claude-3.5-sonnet | 0.21 | 0.96 | 512 | 1,392 | 95 | 1,487 | 1,547 | 0.47584 | 0.351 | true | toxicity | false | true | false | false | false | false | 0 | none | null | no_feedback | true | 1 | add_constraints_to_prompt | auto_resolved | high | I'm stuck on this and need some guidance: Follow-up: the previous reply did not fully solve the issue. Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? Can you walk me through a good solution? | We can make this robust with a few changes. Signal: toxicity while working on login issues. Quick diagnosis: Add a rewrite pathway: sanitize unsafe content into safe alternatives. Recommended... | 961.215255 | 2025-03-07 | 22 | 4 | false | false | 14.6526 | Follow-up: the previous reply did not fully solve the issue. Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | a52fb80193c7c0769c0836852624f4b025a04d2a | 194 | true |
INT008863 | S00146 | U00303 | 2025-04-17T14:01:52Z | train | web_app | customer_support | EG | EMEA | pro | individual | support_strict | meta | llama-3.1-70b | 0.762 | 0.894 | 768 | 219 | 576 | 795 | 1,210 | 0.159 | 0.65 | false | none | false | false | false | false | false | false | 0 | none | -1 | negative | false | 0 | none | auto_resolved | medium | Quick question: A user says they were double charged this month. Help me draft a clear explanation and next steps. I'd like a practical, engineering-focused answer. | This is a common pattern; the fix is straightforward. Goal: handle slow responses reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Add... | 657.024793 | 2025-04-17 | 14 | 3 | false | true | 0.3802 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | d029d0919f8e306d09ca7001d500c1e6f1c81007 | 240 | true |
INT008864 | S01034 | U00021 | 2025-02-25T00:53:40Z | train | web_app | customer_support | AU | APAC | pro | team | creative_writer | openai | gpt-4.1 | 0.494 | 0.853 | 768 | 1,413 | 768 | 2,181 | 1,710 | 0.8724 | 0.404 | true | tool_error | false | false | false | false | true | false | 1 | tool_invocation_failed | 0 | neutral | true | 0 | manual_post_edit | escalated_to_human | medium | Quick question: A user says they were double charged this month. Help me draft a clear explanation and next steps. Please explain step by step. | Below is a decision-focused breakdown. Signal: tool_error while working on slow responses. Quick diagnosis: Validate tool arguments before calling the tool; reject invalid calls early. Recommended actions:... | 1,275.438596 | 2025-02-25 | 0 | 1 | false | false | 1.8398 | A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 7a490bfafd7436aea43a528e9fe4cb6db10b9a6b | 208 | true |
INT008865 | S00732 | U00326 | 2025-04-30T22:01:15Z | test | web_app | data_analysis | FR | EMEA | free | enterprise_team | general_assistant | meta | llama-3.1-8b | 0.23 | 0.95 | 2,048 | 1,161 | 632 | 1,793 | 2,019 | 0.1076 | 0.893 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Can you suggest a few plots to explore churn risk in this dataset of subscriptions? Can you walk me through a good solution? | Let's make the behavior deterministic and observable. Goal: handle safety flag rate reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Add... | 888.063398 | 2025-04-30 | 22 | 2 | false | false | 1.837 | Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | 0af51e7e9a94eeacb4db5e80369428b0b639dc4e | 242 | true |
INT008866 | S00928 | U00216 | 2025-02-21T08:18:15Z | train | web_app | content_writing | IN | APAC | enterprise | team | code_assistant_secure | meta | llama-3.1-8b | 0.735 | 1 | 768 | 60 | 472 | 532 | 737 | 0.03192 | 0.598 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | escalated_to_human | medium | Draft a friendly email announcing a small price increase to our existing customers. | Here's what I'd implement in production. Goal: handle observability metrics reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. -... | 721.845319 | 2025-02-21 | 8 | 4 | false | false | 0.1271 | Draft a friendly email announcing a small price increase to our existing customers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | 20f1dad1d39f7b8df6faaa166fe78024f8d03fd6 | 234 | true |
INT008867 | S00313 | U00434 | 2025-03-24T11:07:59Z | train | mobile_app | coding_assistant | AE | EMEA | pro | team | analytics_helper | openai | gpt-4o-mini | 0.151 | 0.904 | 1,024 | 70 | 360 | 430 | 2,262 | 0.0086 | 0.81 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | medium | I'm stuck on this and need some guidance: Can you refactor this function to be more readable and efficient while keeping the same behaviour? I'd like a practical, engineering-focused answer. | Let's break this down into actionable steps. Goal: handle FastAPI endpoint design reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer simple... | 190.097259 | 2025-03-24 | 11 | 0 | false | true | 0.1944 | Can you refactor this function to be more readable and efficient while keeping the same behaviour? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 54f94ac4f6df7ccc4808605ae4fb839b7a1fbfb2 | 250 | true |
INT008868 | S00727 | U00143 | 2025-04-03T00:14:55Z | test | web_app | customer_support | EG | EMEA | free | team | support_strict | openai | gpt-4o-mini | 0.562 | 0.849 | 768 | 86 | 214 | 300 | 2,318 | 0.006 | 0.556 | true | formatting_error | false | false | false | true | false | false | 1 | vector_search | -2 | negative | false | 1 | regenerate_with_lower_temp | auto_resolved | high | Context for you: Follow-up: the previous reply did not fully solve the issue. Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? Can you walk me through a good solution? | Let's break this down into actionable steps. Signal: formatting_error while working on data export request. Quick diagnosis: Use constrained generation (schema-first prompt) and add unit tests for... | 129.421915 | 2025-04-03 | 0 | 3 | false | false | 0.4019 | Follow-up: the previous reply did not fully solve the issue. Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | d029d0919f8e306d09ca7001d500c1e6f1c81007 | 199 | true |
INT008869 | S00854 | U00300 | 2025-02-27T17:54:50Z | train | web_app | coding_assistant | GB | EMEA | pro | team | creative_writer | anthropic | claude-3.5-sonnet | 0.365 | 0.892 | 512 | 67 | 435 | 502 | 503 | 0.16064 | 0.934 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Quick question: Can you refactor this function to be more readable and efficient while keeping the same behaviour? A short, clear answer would be ideal. | Here's the clean way to approach it. Goal: handle JSON schema validation reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Prefer... | 998.011928 | 2025-02-27 | 17 | 3 | false | true | 0.154 | Can you refactor this function to be more readable and efficient while keeping the same behaviour? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 83c939e9b891e505ac06548e158556e919317a64 | 238 | true |
INT008870 | S00502 | U00089 | 2025-02-27T21:04:37Z | val | api | coding_assistant | BR | AMER | pro | team | internal_policy_qa | anthropic | claude-3.5-sonnet | 0.191 | 0.98 | 1,024 | 1,272 | 927 | 2,199 | 807 | 0.70368 | 0.568 | true | formatting_error | false | false | false | true | false | false | 0 | none | -1 | negative | true | 1 | regenerate_with_lower_temp | abandoned | low | Context for you: Follow-up: the previous reply did not fully solve the issue. I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | Below is a decision-focused breakdown. Let's stabilize it. Signal: formatting_error while working on unit test mocking. Quick diagnosis: Use constrained generation (schema-first prompt) and add unit tests for formatting. Recommended... | 2,724.907063 | 2025-02-27 | 21 | 3 | false | false | 1.3722 | Follow-up: the previous reply did not fully solve the issue. I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | f84eff1764fee544ced395082209fcd5e43e6a0c | 235 | true |
INT008871 | S01008 | U00377 | 2025-03-19T10:50:48Z | train | slack | customer_support | GB | EMEA | free | team | general_assistant | google | gemini-1.5-pro | 0.762 | 0.915 | 1,536 | 629 | 27 | 656 | 1,434 | 0.20992 | 0.505 | true | formatting_error | false | false | false | true | false | false | 2 | code_run|email_send | null | no_feedback | true | 1 | switch_model_family | escalated_to_human | low | Context for you: Follow-up: the previous reply did not fully solve the issue. A user says they were double charged this month. Help me draft a clear explanation and next steps. Please explain step by step. | Below is a practical plan you can execute. Signal: formatting_error while working on login issues. Quick diagnosis: Use constrained generation (schema-first prompt) and add unit tests for formatting.... | 457.461646 | 2025-03-19 | 10 | 2 | false | true | 23.2963 | Follow-up: the previous reply did not fully solve the issue. A user says they were double charged this month. Help me draft a clear explanation and next steps. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | a52fb80193c7c0769c0836852624f4b025a04d2a | 202 | true |
INT008872 | S01389 | U00129 | 2025-04-14T13:52:10Z | train | web_app | internal_qa | FR | EMEA | free | individual | creative_writer | openai | gpt-4o-mini | 0.532 | 0.929 | 2,048 | 780 | 258 | 1,038 | 1,066 | 0.02076 | 0.495 | true | safety_block | false | false | true | false | false | false | 0 | none | null | no_feedback | true | 0 | regenerate_with_lower_temp | escalated_to_human | high | Summarise this internal security policy into three bullet points for non-technical employees. I'd like a practical, engineering-focused answer. | Below is a decision-focused breakdown. Signal: safety_block while working on provider degradation. Quick diagnosis: Route to a safe-help template and suggest benign resources or best practices. Recommended... | 973.733583 | 2025-04-14 | 13 | 0 | false | true | 3.0233 | Summarise this internal security policy into three bullet points for non-technical employees. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | 3ee501c0eea07616af572d79e74d4f247c4b4851 | 208 | true |
INT008873 | S00406 | U00386 | 2025-02-19T07:48:31Z | val | web_app | data_analysis | IN | APAC | free | individual | analytics_helper | meta | llama-3.1-8b | 0.186 | 0.995 | 768 | 1,016 | 161 | 1,177 | 902 | 0.07064 | 0.23 | true | hallucination | true | false | false | false | false | false | 0 | none | null | no_feedback | false | 1 | decompose_task | auto_resolved | low | I'm working on something and need help: Follow-up: the previous reply did not fully solve the issue. I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. Please explain step by step. | We can make this robust with a few changes. Let's stabilize it. Signal: hallucination while working on user satisfaction. Quick diagnosis: Use a 'claim -> evidence -> answer' structure and log missing evidence. Recommended actions: -... | 1,304.878049 | 2025-02-19 | 7 | 2 | false | false | 6.3106 | Follow-up: the previous reply did not fully solve the issue. I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | fbaac57cdb64395ff60fbf79b148e22caaff1431 | 236 | true |
INT008874 | S01289 | U00229 | 2025-02-04T19:21:17Z | train | api | coding_assistant | US | AMER | free | team | internal_policy_qa | openai | gpt-4o-mini | 0.146 | 0.982 | 1,536 | 819 | 434 | 1,253 | 831 | 0.02504 | 0.461 | true | safety_block | false | false | true | false | false | false | 1 | code_run | -2 | negative | false | 0 | enable_tools | auto_resolved | low | Quick question: Write a small example that shows how to handle exceptions properly in Python. | Here's what I'd implement in production. Signal: safety_block while working on rate limiting. Quick diagnosis: Log the trigger category to tune your safety policy. Recommended actions: - Implement retry with... | 1,507.821901 | 2025-02-04 | 19 | 1 | false | false | 1.8871 | Write a small example that shows how to handle exceptions properly in Python. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | f84eff1764fee544ced395082209fcd5e43e6a0c | 210 | true |
INT008875 | S01221 | U00113 | 2025-03-14T08:10:49Z | val | internal_tool | customer_support | GB | EMEA | pro | individual | analytics_helper | local | custom-local-8b | 0.918 | 0.876 | 768 | 626 | 768 | 1,394 | 983 | 0.00556 | 0.468 | true | hallucination | true | false | false | false | false | false | 0 | none | -2 | negative | false | 0 | manual_post_edit | abandoned | medium | Can you help me debug this? Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | Let's break this down into actionable steps. Signal: hallucination while working on billing dispute. Quick diagnosis: Refuse when confidence is low; avoid confident guesses. Recommended actions: - Introduce... | 1,418.107833 | 2025-03-14 | 8 | 4 | false | false | 0.8151 | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 598eba0ae15195966a9835b68c542f0b34fb88e4 | 209 | true |
INT008876 | S00366 | U00238 | 2025-02-20T17:06:05Z | train | api | coding_assistant | EG | EMEA | enterprise | team | support_strict | meta | llama-3.1-70b | 0.233 | 0.904 | 768 | 749 | 768 | 1,517 | 1,270 | 0.3034 | 0.973 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | escalated_to_human | low | I'm stuck on this and need some guidance: I'm getting a Go error: `NullPointerException`. Can you explain what it means and how to fix it? Can you walk me through a good solution? | Let's make the behavior deterministic and observable. Goal: handle FastAPI endpoint design reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure... | 1,194.488189 | 2025-02-20 | 17 | 3 | false | true | 0.9753 | I'm getting a Go error: `NullPointerException`. Can you explain what it means and how to fix it? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | da34aac46ee8d6d44012043be92e33884c6b4920 | 234 | true |
INT008877 | S00606 | U00171 | 2025-04-02T13:02:26Z | val | web_app | content_writing | GB | EMEA | free | team | internal_policy_qa | google | gemini-1.5-pro | 1.113 | 0.909 | 512 | 616 | 248 | 864 | 1,238 | 0.27648 | 0.602 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | I'm working on something and need help: Write a concise product description for a new wireless mouse aimed at software developers. A short, clear answer would be ideal. | Below is a decision-focused breakdown. Goal: handle A/B testing reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. -... | 697.899838 | 2025-04-02 | 13 | 2 | false | true | 2.4839 | Write a concise product description for a new wireless mouse aimed at software developers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | db47a2ba00ac8f8120dcd2fd47d3648a8f594c66 | 217 | true |
INT008878 | S00725 | U00362 | 2025-04-18T11:02:11Z | test | internal_tool | customer_support | FR | EMEA | free | individual | analytics_helper | openai | gpt-4.1 | 0.686 | 0.862 | 1,024 | 1,194 | 148 | 1,342 | 1,907 | 0.5368 | 0.862 | false | none | false | false | false | false | false | false | 1 | knowledge_lookup | null | no_feedback | false | 0 | none | escalated_to_human | low | Can you help me debug this? Our customer is getting an error when trying to log in to their account. Can you help me debug this? Please explain step by step. | I'll keep this concrete and measurable. Goal: handle missing feature reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Prefer simple... | 703.723125 | 2025-04-18 | 11 | 4 | false | true | 8.0676 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 598eba0ae15195966a9835b68c542f0b34fb88e4 | 241 | true |
INT008879 | S00512 | U00389 | 2025-04-17T12:54:07Z | train | web_app | data_analysis | EG | EMEA | free | individual | creative_writer | local | custom-local-8b | 0.141 | 0.933 | 256 | 832 | 256 | 1,088 | 826 | 0.00436 | 0.511 | true | formatting_error | false | false | false | true | false | false | 3 | code_run|sql_query|workflow_orchestrator | 0 | neutral | true | 2 | regenerate_with_lower_temp | auto_resolved | low | Follow-up: the previous reply did not fully solve the issue. Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | We can make this robust with a few changes. Signal: formatting_error while working on latency p95. Quick diagnosis: Your output likely violated the contract. Enforce schema validation and... | 1,317.191283 | 2025-04-17 | 12 | 3 | false | true | 3.25 | Follow-up: the previous reply did not fully solve the issue. Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 814335a175319858ca540d38bfcac020ec643c3f | 190 | true |
INT008880 | S00561 | U00197 | 2025-02-24T12:55:49Z | val | api | coding_assistant | DE | EMEA | free | enterprise_team | creative_writer | openai | gpt-4o-mini | 0.354 | 0.944 | 1,536 | 484 | 432 | 916 | 2,254 | 0.01832 | 0.902 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Quick question: I'm getting a Go error: `IndexError`. Can you explain what it means and how to fix it? Can you walk me through a good solution? | This is a common pattern; the fix is straightforward. Goal: handle API retry logic reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for API retry logic and add tests before optimizing. - Add... | 406.388642 | 2025-02-24 | 12 | 0 | false | true | 1.1204 | I'm getting a Go error: `IndexError`. Can you explain what it means and how to fix it? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 83c939e9b891e505ac06548e158556e919317a64 | 258 | true |
INT008881 | S00211 | U00013 | 2025-03-31T13:34:11Z | test | web_app | content_writing | BR | AMER | pro | enterprise_team | creative_writer | anthropic | claude-3.5-sonnet | 0.807 | 0.964 | 1,024 | 1,417 | 955 | 2,372 | 1,504 | 0.75904 | 0.875 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | high | I'm stuck on this and need some guidance: Draft a friendly email announcing a small price increase to our existing customers. A short, clear answer would be ideal. | Let's turn this into a reliable workflow. Goal: handle observability metrics reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure... | 1,577.12766 | 2025-03-31 | 13 | 0 | false | true | 1.4838 | Draft a friendly email announcing a small price increase to our existing customers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | 469eb15f8b08d9b4130e14a6c3888f14e335a854 | 220 | true |
INT008882 | S00384 | U00259 | 2025-04-20T12:00:50Z | train | slack | data_analysis | IN | APAC | free | enterprise_team | internal_policy_qa | openai | gpt-4.1 | 0.122 | 0.925 | 1,024 | 445 | 175 | 620 | 979 | 0.248 | 0.695 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | high | I'm stuck on this and need some guidance: Help me compute the conversion rate per marketing channel and highlight the top 3 performers. Please explain step by step. | Let's make the behavior deterministic and observable. Goal: handle error rate reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for error rate and add tests before... | 633.299285 | 2025-04-20 | 12 | 6 | true | true | 2.5429 | Help me compute the conversion rate per marketing channel and highlight the top 3 performers. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 62646e5a7026d4f68ab1867a730451326aa3b664 | 230 | true |
INT008883 | S00315 | U00048 | 2025-03-16T16:45:22Z | train | web_app | customer_support | IN | APAC | free | enterprise_team | creative_writer | meta | llama-3.1-70b | 0.445 | 0.966 | 768 | 1,125 | 155 | 1,280 | 451 | 0.256 | 0.661 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | Our customer is getting an error when trying to log in to their account. Can you help me debug this? Please explain step by step. | We can make this robust with a few changes. Goal: handle missing feature reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for missing feature and add tests before... | 2,838.137472 | 2025-03-16 | 16 | 6 | true | true | 7.2581 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 7a490bfafd7436aea43a528e9fe4cb6db10b9a6b | 230 | true |
INT008884 | S00089 | U00070 | 2025-03-24T21:28:14Z | test | web_app | customer_support | US | AMER | free | enterprise_team | internal_policy_qa | openai | gpt-4.1 | 0.491 | 0.883 | 768 | 522 | 489 | 1,011 | 1,090 | 0.4044 | 0.632 | false | none | false | false | false | false | false | false | 2 | knowledge_lookup|workflow_orchestrator | null | no_feedback | false | 0 | none | auto_resolved | high | I'm stuck on this and need some guidance: I need a short, friendly reply to a customer who is asking about the status of their refund. Please explain step by step. | Here's the clean way to approach it. Goal: handle slow responses reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Start by... | 927.522936 | 2025-03-24 | 21 | 0 | false | false | 1.0675 | I need a short, friendly reply to a customer who is asking about the status of their refund. | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | false | 03bf57d7adc83026d525c8192379aa203b71e1eb | 227 | true |
INT008885 | S00823 | U00325 | 2025-02-19T23:24:55Z | train | mobile_app | coding_assistant | BR | AMER | enterprise | individual | code_assistant_secure | meta | llama-3.1-8b | 0.175 | 0.95 | 1,024 | 1,237 | 423 | 1,660 | 499 | 0.0996 | 0.77 | false | none | false | false | false | false | false | false | 0 | none | 1 | positive | false | 0 | none | auto_resolved | high | Context for you: I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. I'd like a practical, engineering-focused answer. | Here's what I'd implement in production. Goal: handle unit test mocking reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. - Start by writing a... | 3,326.653307 | 2025-02-19 | 23 | 2 | false | false | 2.9243 | I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | 52eadb52b3319d2dab569ce8fb653a0bee56a90f | 244 | true |
INT008886 | S01499 | U00076 | 2025-04-13T21:38:12Z | val | api | coding_assistant | IN | APAC | free | enterprise_team | internal_policy_qa | local | custom-local-8b | 0.21 | 0.911 | 512 | 549 | 390 | 939 | 1,566 | 0.00376 | 0.645 | true | safety_block | false | false | true | false | false | false | 3 | code_run|vector_search|workflow_orchestrator | null | no_feedback | true | 0 | regenerate_with_lower_temp | abandoned | low | Context for you: I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. A short, clear answer would be ideal. | Here's what I'd implement in production. Signal: safety_block while working on API retry logic. Quick diagnosis: Route to a safe-help template and suggest benign resources or best practices. Recommended... | 599.616858 | 2025-04-13 | 21 | 6 | true | false | 1.4077 | I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | f84eff1764fee544ced395082209fcd5e43e6a0c | 205 | true |
INT008887 | S01452 | U00044 | 2025-04-20T01:38:46Z | train | api | data_analysis | BR | AMER | free | individual | code_assistant_secure | openai | gpt-4o-mini | 0.202 | 0.907 | 1,024 | 771 | 123 | 894 | 1,206 | 0.01788 | 0.53 | true | hallucination | true | false | false | false | false | false | 0 | none | -2 | negative | true | 1 | regenerate_with_lower_temp | escalated_to_human | medium | Follow-up: the previous reply did not fully solve the issue. Can you suggest a few plots to explore churn risk in this dataset of subscriptions? I'd like a practical, engineering-focused answer. | This is a common pattern; the fix is straightforward. Signal: hallucination while working on safety flag rate. Quick diagnosis: Refuse when confidence is low; avoid confident guesses. Recommended actions:... | 741.293532 | 2025-04-20 | 1 | 6 | true | false | 6.2683 | Follow-up: the previous reply did not fully solve the issue. Can you suggest a few plots to explore churn risk in this dataset of subscriptions? | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 4e275efe35fdc513983e7d3f3ad32b50cc391969 | 207 | true |
INT008888 | S00746 | U00192 | 2025-02-25T02:10:43Z | train | web_app | internal_qa | AU | APAC | free | enterprise_team | analytics_helper | openai | gpt-4.1 | 0.603 | 0.917 | 1,536 | 468 | 424 | 892 | 1,549 | 0.3568 | 0.98 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | I'm stuck on this and need some guidance: Explain our refund policy in simple language that a new support agent can remember. Can you walk me through a good solution? | I'll keep this concrete and measurable. Goal: handle errors reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer simple baselines first;... | 575.855391 | 2025-02-25 | 2 | 1 | false | false | 1.1038 | Explain our refund policy in simple language that a new support agent can remember. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | d33eca392020f2e6e14ec7fe3091a25b46d6bae1 | 245 | true |
INT008889 | S00812 | U00030 | 2025-04-09T09:19:20Z | test | api | internal_qa | IN | APAC | free | individual | internal_policy_qa | anthropic | claude-3.5-sonnet | 0.681 | 1 | 768 | 155 | 476 | 631 | 1,235 | 0.20192 | 0.503 | true | latency_timeout | false | false | false | false | false | true | 0 | none | null | no_feedback | false | 2 | regenerate_with_lower_temp | escalated_to_human | critical | Context for you: Follow-up: the previous reply did not fully solve the issue. Explain our refund policy in simple language that a new support agent can remember. I'd like a practical, engineering-focused answer. | Let's break this down into actionable steps. Signal: latency_timeout while working on latency. Quick diagnosis: Reduce payload size and enable caching for repeated queries. Recommended actions: -... | 510.931174 | 2025-04-09 | 9 | 2 | false | true | 0.3256 | Follow-up: the previous reply did not fully solve the issue. Explain our refund policy in simple language that a new support agent can remember. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | false | 8ad879fd7895b38236ec6ff25a956251dc06e51b | 198 | true |
INT008890 | S01247 | U00358 | 2025-02-03T12:32:58Z | train | web_app | coding_assistant | FR | EMEA | pro | individual | internal_policy_qa | meta | llama-3.1-70b | 0.209 | 0.931 | 1,024 | 1,484 | 705 | 2,189 | 814 | 0.4378 | 0.953 | false | none | false | false | false | false | false | false | 2 | knowledge_lookup|sql_query | 2 | positive | false | 0 | none | auto_resolved | medium | Can you refactor this function to be more readable and efficient while keeping the same behaviour? Can you walk me through a good solution? | Below is a decision-focused breakdown. Goal: handle unit test mocking reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions... | 2,689.189189 | 2025-02-03 | 12 | 0 | false | true | 2.105 | Can you refactor this function to be more readable and efficient while keeping the same behaviour? | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | true | f84eff1764fee544ced395082209fcd5e43e6a0c | 217 | true |
INT008891 | S01330 | U00214 | 2025-02-16T16:28:46Z | train | api | coding_assistant | GB | EMEA | enterprise | team | creative_writer | openai | gpt-4.1 | 0.223 | 0.851 | 512 | 657 | 255 | 912 | 1,489 | 0.3648 | 0.818 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | medium | I'm working on something and need help: I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. I'd like a practical, engineering-focused answer. | Let's turn this into a reliable workflow. Goal: handle FastAPI endpoint design reliably, with guardrails for formatting and failure handling. Plan: - Prefer simple baselines first; then iterate based on measured failure slices. -... | 612.491605 | 2025-02-16 | 16 | 6 | true | true | 2.5765 | I need help understanding why this asynchronous code is not awaiting correctly. Explain with a small code snippet. | You are a senior software engineer helping users with code, debugging, and best practices. Provide clear explanations, structured steps, and high-quality code examples when relevant. | false | 83c939e9b891e505ac06548e158556e919317a64 | 232 | true |
INT008892 | S01393 | U00036 | 2025-04-04T11:26:24Z | train | internal_tool | data_analysis | EG | EMEA | free | individual | creative_writer | meta | llama-3.1-8b | 0.357 | 0.868 | 2,048 | 924 | 396 | 1,320 | 1,402 | 0.0792 | 0.821 | false | none | false | false | false | false | false | false | 1 | workflow_orchestrator | null | no_feedback | false | 0 | none | abandoned | low | I'm stuck on this and need some guidance: Help me compute the conversion rate per marketing channel and highlight the top 3 performers. A short, clear answer would be ideal. | Let's make the behavior deterministic and observable. Goal: handle tool-call success rate reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for tool-call success rate... | 941.512126 | 2025-04-04 | 11 | 4 | false | true | 2.3333 | Help me compute the conversion rate per marketing channel and highlight the top 3 performers. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | 814335a175319858ca540d38bfcac020ec643c3f | 233 | true |
INT008893 | S01052 | U00012 | 2025-03-29T10:35:55Z | train | internal_tool | customer_support | AU | APAC | free | team | internal_policy_qa | google | gemini-1.5-pro | 0.517 | 0.822 | 2,048 | 296 | 377 | 673 | 1,807 | 0.21536 | 0.969 | false | none | false | false | false | false | false | false | 2 | http_request|workflow_orchestrator | 0 | neutral | false | 0 | none | auto_resolved | medium | Can you help me debug this? Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | Let's turn this into a reliable workflow. Goal: handle unexpected charge reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Start by writing a... | 372.440509 | 2025-03-29 | 10 | 5 | true | true | 0.7851 | Hi, I'm having an issue with my billing. The invoice total doesn't match what I expected. Can you check it? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 03bf57d7adc83026d525c8192379aa203b71e1eb | 246 | true |
INT008894 | S00499 | U00345 | 2025-04-18T08:07:27Z | train | web_app | internal_qa | EG | EMEA | free | individual | creative_writer | openai | gpt-4o-mini | 0.592 | 0.936 | 1,536 | 362 | 499 | 861 | 558 | 0.01724 | 0.821 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | abandoned | high | I'm working on something and need help: Summarise this internal security policy into three bullet points for non-technical employees. A short, clear answer would be ideal. | Let's break this down into actionable steps. Goal: handle safety flags reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions quickly. - Prefer... | 1,543.010753 | 2025-04-18 | 8 | 4 | false | false | 0.7255 | Summarise this internal security policy into three bullet points for non-technical employees. | You are an internal knowledge assistant helping teammates navigate documentation, policies, and internal tools. Be precise, concise, and point to relevant information and next steps. | true | 3ee501c0eea07616af572d79e74d4f247c4b4851 | 236 | true |
INT008895 | S00625 | U00318 | 2025-02-04T07:05:04Z | val | mobile_app | data_analysis | AU | APAC | enterprise | enterprise_team | general_assistant | meta | llama-3.1-8b | 0.22 | 0.929 | 1,536 | 1,432 | 690 | 2,122 | 1,667 | 0.12732 | 0.735 | false | none | false | false | false | false | false | false | 2 | email_send|workflow_orchestrator | 2 | positive | false | 0 | none | abandoned | medium | Quick question: I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. I'd like a practical, engineering-focused answer. | Below is a decision-focused breakdown. Goal: handle user satisfaction reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for user satisfaction and add... | 1,272.945411 | 2025-02-04 | 7 | 1 | false | false | 2.0754 | I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | true | 0af51e7e9a94eeacb4db5e80369428b0b639dc4e | 216 | true |
INT008896 | S00714 | U00160 | 2025-03-02T20:44:32Z | val | mobile_app | customer_support | EG | EMEA | enterprise | team | creative_writer | meta | llama-3.1-70b | 0.68 | 0.891 | 256 | 651 | 256 | 907 | 880 | 0.1814 | 0.583 | true | latency_timeout | false | false | false | false | false | true | 1 | http_request | 0 | neutral | false | 0 | regenerate_with_lower_temp | escalated_to_human | high | I'm working on something and need help: Our customer is getting an error when trying to log in to their account. Can you help me debug this? | Here are the guardrails I'd put in place. Signal: latency_timeout while working on account verification. Quick diagnosis: Reduce payload size and enable caching for repeated queries.... | 1,030.681818 | 2025-03-02 | 20 | 6 | true | false | 2.543 | Our customer is getting an error when trying to log in to their account. Can you help me debug this? | You are a helpful customer support assistant for a SaaS product. Answer clearly, empathetically, and step-by-step while keeping the response concise and practical. | true | 7a490bfafd7436aea43a528e9fe4cb6db10b9a6b | 185 | true |
INT008897 | S00127 | U00379 | 2025-03-06T15:38:13Z | train | slack | content_writing | BR | AMER | pro | individual | support_strict | meta | llama-3.1-70b | 0.968 | 0.899 | 1,024 | 248 | 978 | 1,226 | 865 | 0.2452 | 0.741 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | auto_resolved | low | I'm stuck on this and need some guidance: Write a concise product description for a new wireless mouse aimed at software developers. | Let's make the behavior deterministic and observable. Goal: handle A/B testing reliably, with guardrails for formatting and failure handling. Plan: - Use schema validation (e.g., Pydantic) to prevent bad inputs from propagating. - Prefer simple... | 1,417.34104 | 2025-03-06 | 15 | 3 | false | true | 0.2536 | Write a concise product description for a new wireless mouse aimed at software developers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | 75b7366387b6cf4f5631ec9dd298a359e5bb9c1d | 247 | true |
INT008898 | S00069 | U00437 | 2025-03-25T10:36:19Z | train | api | content_writing | DE | EMEA | free | individual | support_strict | openai | gpt-4o-mini | 0.817 | 0.916 | 768 | 71 | 16 | 87 | 759 | 0.00176 | 0.838 | false | none | false | false | false | false | false | false | 0 | none | 2 | positive | false | 0 | none | auto_resolved | high | Context for you: Draft a friendly email announcing a small price increase to our existing customers. I'd like a practical, engineering-focused answer. | Here's a clear, step-by-step answer. Goal: handle LLM reliability reliably, with guardrails for formatting and failure handling. Plan: - Add telemetry fields for latency/tokens/cost so you can debug regressions... | 114.624506 | 2025-03-25 | 10 | 1 | false | true | 4.4375 | Draft a friendly email announcing a small price increase to our existing customers. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | true | 75b7366387b6cf4f5631ec9dd298a359e5bb9c1d | 213 | true |
INT008899 | S01325 | U00016 | 2025-03-12T15:17:28Z | train | mobile_app | data_analysis | GB | EMEA | free | individual | internal_policy_qa | google | gemini-1.5-pro | 0.376 | 0.991 | 1,024 | 896 | 239 | 1,135 | 2,062 | 0.3632 | 0.89 | false | none | false | false | false | false | false | false | 2 | email_send|sql_query | null | no_feedback | false | 0 | none | escalated_to_human | medium | I'm stuck on this and need some guidance: I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | Let's turn this into a reliable workflow. Goal: handle token usage reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for token usage and add tests before... | 550.436469 | 2025-03-12 | 15 | 2 | false | true | 3.749 | I have a CSV with user sessions. Show me a quick breakdown of sessions per country and device type. | You are a data analysis assistant helping users reason about data, metrics, and models. Explain trade-offs, outline steps, and keep the language accessible but precise. | false | 62646e5a7026d4f68ab1867a730451326aa3b664 | 220 | true |
INT008900 | S00718 | U00306 | 2025-04-30T08:28:22Z | train | mobile_app | content_writing | DE | EMEA | pro | team | general_assistant | google | gemini-1.5-pro | 0.721 | 0.904 | 1,536 | 960 | 499 | 1,459 | 1,312 | 0.46688 | 1 | false | none | false | false | false | false | false | false | 0 | none | null | no_feedback | false | 0 | none | escalated_to_human | high | Can you help me debug this? Turn these bullet points into a short, engaging blog introduction about learning machine learning. Can you walk me through a good solution? | Below is a practical plan you can execute. Goal: handle evaluation for LLM systems reliably, with guardrails for formatting and failure handling. Plan: - Start by writing a small function boundary for evaluation for LLM... | 1,112.042683 | 2025-04-30 | 8 | 2 | false | false | 1.9238 | Turn these bullet points into a short, engaging blog introduction about learning machine learning. | You are a writing assistant helping users craft clear, engaging, and well-structured content. Keep the tone natural, readable, and aligned with the user's intent. | false | a1c74d04f735d164fdc621e3ce3d5bfea1026243 | 222 | true |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.