Skip to content

Latest commit

 

History

History
25 lines (23 loc) · 1.22 KB

File metadata and controls

25 lines (23 loc) · 1.22 KB
id write-llm-evals
type playbook
title Write Deterministic Evals for LLMs and Agents
summary Build regression test suites for LLM prompt changes and tool-calling agents with golden datasets and schema assertions.
lang en-US
content_version 1
status reviewed
reviewed_on 2026-09-02

Write Deterministic Evals for LLMs and Agents

  1. Curate a versioned golden dataset of real user inputs covering happy paths, adversarial prompts, ambiguous edge cases, and known regressions.
  2. Define deterministic boundary assertions before adding semantic judges: validate JSON Schema conformity, required fields, and disallowed tokens.
  3. Test tool-calling parameters against strict type contracts; assert that tool selections match expected capabilities without hallucinated arguments.
  4. Separate cheap local unit tests from live model evaluations. Use recorded fixture responses for fast CI runs and run live evals on scheduled batches.
  5. Record benchmark pass rates, token counts, and latency before and after any prompt or model migration. Never ship prompt changes without comparing diffs.
  6. Guard against flaky evals by establishing tolerance thresholds and isolating temperature/seed parameters during regression checks.