AI Prompt for Evals & Observability
Design A/B rollout analysis and drift detection for faithfulness on a production LLM app in summarization feed.
More prompts for Evals & Observability.
Instrument, query, and triage structured extraction LLM app traces in Lunary with TypeScript SDK, covering latency, cost, and quality dashboards.
Instrument, query, and triage classification pipeline LLM app traces in OpenTelemetry + Jaeger with Ruby SDK, covering latency, cost, and quality dashboards.
Instrument, query, and triage agent with tool-use LLM app traces in Galileo with Java SDK, covering latency, cost, and quality dashboards.
Instrument, query, and triage code-completion copilot LLM app traces in OpenTelemetry + Jaeger with Java SDK, covering latency, cost, and quality dashboards.
Design a pairwise + rubric LLM-as-judge prompt for multi-turn dialogue with bias mitigation, calibration, and reproducibility.
Design a pairwise + rubric LLM-as-judge prompt for SQL generation with bias mitigation, calibration, and reproducibility.