- Handle the production lifecycle of ML systems, including training, packaging, deployment, monitoring, and controlled rollouts.
- Design, build, and improve training and retraining pipelines for data ingestion, labeling and QA, feature generation, experiment tracking, and reproducibility.
- Build and improve high-performance inference services and deployment patterns.
- Optimize inference using ONNX/TensorRT, quantization, batching, and GPU/CPU tuning.
- Partner with DevOps/SRE on scaling, reliability, incident readiness, observability, and alerting.
- Own model versioning, CI/CD, and release workflows for reproducible and safe production delivery.
- Strengthen data and model monitoring for performance, drift, and quality, and implement feedback loops.
- Communicate results and trade-offs to stakeholders.
- Evaluate and validate model quality against evolving data and business needs.
AWSDockerPython+4 more