Shield AI · posted · checked
Staff Software Engineer ML Platform (6244)
Python$200k – $300kOn-site · San DiegoStaff+
Who can apply
On-site · San Diego
Location: San Diego, California / San Mateo, California
Shield AI is a venture-backed defense-tech company with the mission of protecting service members and civilians with intelligent systems. Its products include Hivemind autonomy software, V-BAT and X-BAT aircraft, and Aechelon simulation and synthetic reality technologies. With offices and facilities across the U.S., Europe, the Middle East, and Asia-Pacific, Shield AI’s technology actively supports operations worldwide. For more information, visit www.shield.ai. Follow Shield AI on LinkedIn, X, Instagram, and YouTube.
Job Description:
Model development produces experiments, checkpoints, evaluation results and many versions of the same model. Keeping those records connected lets engineers reproduce results, compare approaches and move a trained model into use.
We are looking for a Senior Software Engineer to build the backend services that make those workflows reliable. You will connect training runs to their data, code and configuration, manage model artifacts and help machine-learning engineers spend less time reconstructing what happened in an experiment.
What you'll do:
Build experiment-tracking and model-registry services, integrating tools such as MLflow where appropriate.
Connect exact dataset, code and configuration versions to training runs, checkpoints and model versions.
Develop APIs and adapters that connect training jobs, data services and artifact storage.
Make interrupted or failed workflows recoverable through reliable state handling, retries and useful diagnostics.
Make checkpoints, model artifacts and evaluation results easy to find and retrieve for further training, evaluation or deployment.
Work directly with machine-learning engineers to simplify their workflows and reduce manual record keeping.
Deliver tested services with observability and compatible upgrades across supported cloud and customer-managed environments.
Key Outcomes:
Engineers can trace a model version back to the data, code, configuration and run that produced it.