As a Senior MLOps Engineer, you will have a high level of ownership over the platform the machine learning team builds on, spanning model training, experimentation, deployment, monitoring, and the data infrastructure supporting production ML.
You must have firsthand experience building MLOps infrastructure from the ground up or taking an existing platform through significant scaling and continuing to operate it as it matured.
Responsibilities include:
- Own and extend the ML platform end to end, including training orchestration, experiment tracking, model registry, deployment, and production monitoring.
- Build and operate data pipelines supporting model training and online inference.
- Build backend services and APIs that support the deployment and operation of ML models in production.
- Design processes for backfills, replays, and recovery when upstream data feeds fail.
- Ensure training data accurately reflects what was known at the point in time it represents.
- Build and manage reliable processes for moving trained models into production.
- Ensure model training runs and results are reproducible.
- Monitor deployed model performance over time, including models where outcomes are confirmed later.
- Manage training and inference costs as data volume and the number of production models grow.
- Contribute to technical architecture design and reviews.