Evaluation and Monitoring

Official references for this objective

• Databricks — The 52x Multiplier: How Zepto Mastered AI Agent Evaluation at Scale | Databricks

  establishing “Evaluation as Infrastructure”—covering LLM-as-a-Judge calibration, strategic sampling, and automated feedback loops

• Databricks — Agent as a Judge: Scaling AI Evaluation for the Agentic Era | Databricks

  running on Phoenix’s own production traces

• Databricks — Stop Guessing, Start Scoring: LLM Extraction Evaluation on Databricks | Databricks

  we use managed MLflow, custom scorers, and a Databricks App with Lakebase to capture SME feedback and structured ground truth