Systems I build, ship, and keep running with strong feedback loops.
The complete data pipeline: CSV exports are ingested by deejay-cog, normalized, and posted to api-kaianolevine-com — a FastAPI service backed by PostgreSQL on Railway. Exposes a typed REST API serving 10+ years of set history. Feature flags control activation without deployment, managed via API endpoint.
Python cogs built on common-python-utils: watcher-cog polls Google Drive every minute and fires Prefect flow runs, deejay-cog processes DJ set CSVs, evaluator-cog runs AI conformance checks post-pipeline. All use uv, ruff, pytest, and semantic-release CI.
After each pipeline run, Claude evaluates conformance against the ecosystem standards document and writes structured findings to PostgreSQL. Findings are live — see Pipeline Health for current system state.