Wiring Replay Evals Into a CI/CD Pipeline for LLM Agents
Freeze production traces to test LLM agent changes without rerunning the entire system.
Estelle Zola
Contributing Editor
Estelle Zola
Estelle Zola is a contributing editor at Ship Signals covering replay evaluation design. Based in Cape Town, Estelle has written for Ship Signals since 2015.
1 story · Cape Town