Senior Machine Learning Engineer

Role overview Own the reliability, scalability, and automation of production machine-learning pipelines, spanning cloud infrastructure and the model lifecycle. You’ll improve the systems that prepare data, train models, and generate predictions, taking platform changes from design through deployment while keeping live services dependable.

Responsibilities - Package pipeline services in Docker and run them on cloud infrastructure, designing scalable and event-driven execution for data preparation, training, and prediction workloads. - Build and maintain automated build, test, and deployment workflows, and improve production visibility through logs, metrics, and alerts. - Make training environments reproducible and strengthen experiment tracking, model versioning, management, and deployment. - Modernize pipeline architecture incrementally, coordinating with data science and DevOps colleagues and avoiding disruption to live systems. - Improve infrastructure efficiency by selecting suitable compute and storage resources.

Requirements - At least five years operating production Python systems, with sound practices in testing, version control, code review, and CI/CD. - Practical Docker experience, including optimizing Dockerfiles, using multi-stage builds, and troubleshooting containers in production. - Cloud experience with compute and storage services; the existing environment includes AWS EC2, ECS, S3, EFS, CloudWatch, and IAM. - Experience building automated pipelines with GitHub Actions or an equivalent tool. - Working knowledge of document and relational databases, such as MongoDB and PostgreSQL, including secure connections from containerized services. - Production ML lifecycle experience, including reproducible training, experiment tracking and model management with MLflow or a comparable tool, and model deployment.

Nice to have - Python environment and configuration tools such as uv, pydantic, or Hydra/OmegaConf. - Workflow orchestration experience with tools such as SageMaker Pipelines or Prefect. - Familiarity with ML libraries including TensorFlow, scikit-learn, or LightGBM, or with geospatial data and GIS tools such as geopandas, PostGIS, or ArcGIS.

Benefits and work setup - Full-time role with end-to-end ownership of platform initiatives and collaboration across engineering and data science.