About this role
Own and operate the end-to-end development loop for humans and AI agents, including fast reproducible environments, deterministic CI, agent harness tooling and orchestration, evaluation-gated merges, and measurement to drive recursive harness and model improvements.
Skills for this role
PythonContainersKubernetesCI/CD systemsDeterministic CIAI coding agentsClaude CodeCursorCopilotAgent harnessesEvaluation suitesEvaluation methodsReinforcement learning from execution feedbackMeasurement and instrumentationBazelBuckNixMonorepo toolingGoRustGit-at-scaleLangSmithLang GraphPostgreSQLpgvectorClickHouseS3-compatible object storageNeo4j EnterpriseOpenTelemetryGrafana