Senior Python Engineer Data Platform · Remote View

Take the models from research to production and keep them there. You'll build the pipelines that assemble training data, the services that serve predictions during a live slate, and the grading jobs that score every pick against the final box score — then make all of it reliable enough that nobody has to babysit it on a Sunday.

• Build production data pipelines that land raw feeds in S3 and modeled tables in Postgres on a schedule.
• Ship model-serving containers to Cloud Run and ECS Fargate with predictable latency under slate-day load.
• Serve precomputed features and predictions from DynamoDB where the app needs single-digit-millisecond reads.
• Own the grading jobs that settle every logged bet against the official result, and the writes that surface the result in the app.
• Set the engineering standard for the data codebase: typing, tests, and reproducible runs.

• Senior-level Python with real production ownership, not notebook-only experience.
• Experience with orchestration (Airflow, Prefect, or similar) and with idempotent, retryable jobs.
• Strong SQL and comfort designing analytical schemas in Postgres.
• Practical AWS and GCP experience — S3, DynamoDB, and Fargate on one side, Cloud Run, Cloud SQL, and Pub/Sub on the other.
• A track record of taking someone else's model and making it run every day without drama.

• You have been on call for a pipeline that had to be right by kickoff.
• Experience with a feature store, or with the reasons you chose not to use one.
• Comfort pairing with a data scientist and turning a notebook into a service without rewriting it twice.