ChatGPT Prompt for Evals & Observability
Design a pairwise + rubric LLM-as-judge prompt for SQL generation with bias mitigation, calibration, and reproducibility.
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.
Instrument, query, and triage classification pipeline LLM app traces in Langfuse with Python SDK, covering latency, cost, and quality dashboards.
Replace the bracketed placeholders with your own context before running the prompt:
[Additional task-specific dimension for SQL generation]— fill in your specific additional task-specific dimension for sql generation.[REFERENCE ANSWER (optional): {reference}]— fill in your specific reference answer (optional): {reference}.["specific problem 1", "specific problem 2"]— fill in your specific "specific problem 1", "specific problem 2".