Summary

This data scientist role owns the measurement and improvement of Lovable's AI agent through evaluation systems, telemetry analysis, and experimentation. You will work closely with agent engineers using SQL, Python, Braintrust, BigQuery, PubSub, and GCP; the working arrangement is not specified.

Responsibilities

• Define metrics for agent quality
• Build evaluation and experiment systems
• Design a/b tests for agent changes
• Analyze agent traces and telemetry
• Develop continuous evaluation tooling
• Establish standards for agent behavior