AI Reliability Console
Monitor behavior, manage risk, and ship confidently
Evaluation Studio
Runs
Browse evaluation runs and inspect activity trends by agent and filter scope.
Evaluation Activity
Daily evaluation volume for the selected filters. Hover a cell for type and count.
All Runs
7 results| Run | Status | Decision | Confidence | Derived Metrics | Decision Metrics | Intents | Details |
|---|---|---|---|---|---|---|---|
Support Assistant v3 - Pre-release Support Assistant · gpt-4.1 | Awaiting Human | Pending | 68% | Helpfulness: 82 Information Density: 79 Format Adherence: 94 | Release Gate Decision: Pending Critical Failure Rate: 2.7% Pass Rate: 89.2% | Release GateRegressionCost Efficiency | |
Concierge Agent - Weekly Regression Concierge Agent · gpt-4o | Completed | Ship | 83% | Helpfulness: 88 Information Density: 85 Format Adherence: 96 | Regression Regression %: 0.9% Result Variance %: 1.6% Pass Rate: 93.6% | RegressionQuality Diagnosis | |
Voice QA - Escalation Sweep Voice QA Agent · gpt-4o-mini | Running | Pending | 52% | Helpfulness: 73 Information Density: 70 Format Adherence: 91 | Release Gate Decision: Pending Critical Failure Rate: 3.4% Pass Rate: 86.9% | Release GateQuality Diagnosis | |
FAQ Assistant - A/B Candidate FAQ Assistant · gpt-4.1-mini | Published | Ship | 88% | Helpfulness: 91 Information Density: 88 Format Adherence: 97 | A/B Model Test Chosen Intent: Release Gate Winner: Candidate B Pass Rate (Winner): 95.1% | A/B Model TestCost Efficiency | |
Support Assistant - Emergency Replay Support Assistant · gpt-4.1 | Failed | No Ship | 91% | Helpfulness: 66 Information Density: 61 Format Adherence: 88 | Release Gate Decision: No Ship Critical Failure Rate: 5.4% Pass Rate: 81.5% | Release GateRegression | |
Concierge Agent - Baseline Snapshot Concierge Agent · gpt-4o | Completed | Ship | 77% | Helpfulness: 83 Information Density: 80 Format Adherence: 94 | Regression Regression %: 1.4% Result Variance %: 2.3% Pass Rate: 90.8% | RegressionQuality Diagnosis | |
FAQ Assistant - Baseline Candidate A FAQ Assistant · gpt-4.1-mini | Completed | Ship | 80% | Helpfulness: 86 Information Density: 84 Format Adherence: 95 | A/B Model Test Chosen Intent: Release Gate Winner: Candidate B Pass Rate (Candidate A): 91.9% | A/B Model TestCost Efficiency |