Staff Engineer, Data Platform (R5659)

Job Description:

Reproducing a model’s results starts with knowing exactly which data it used. Shield AI’s AI Factory gives machine-learning engineers tools to find, access and reuse training data across internal programs and customer deployments.

We are looking for a Staff Engineer to build the services behind those workflows. You will design and implement data APIs, dataset versioning, search and metadata systems that remain useful as datasets grow and schemas change. This role combines technical leadership with substantial hands-on software development.

What you'll do:

• Build dataset discovery, catalog and retrieval services, including LanceDB integration for indexing and search.

• Create APIs that let developers pin an exact dataset version and retrieve only the data they are authorized to access.

• Track where data came from and how it changed so engineers can understand and reproduce training results.

• Design schema validation, compatibility and migration mechanisms that preserve access to historical datasets.

• Define common identifiers and metadata relationships that connect datasets to experiments and model artifacts.

• Work with machine-learning engineers to understand how they create and use datasets, then turn recurring needs into reusable tools.

• Deliver reliable data services with tested upgrades, clear diagnostics and support for large datasets.

Key outcomes:

• Developers can find and retrieve the right dataset without assembling a new data pipeline for each experiment.

• Training runs can reference exact dataset versions and trace them back to their sources.

• Schema changes have explicit compatibility or migration paths for existing users.

• Data services work across multiple model-development workflows and deployment environments.

Required qualifications:

• Experience designing and implementing production data platforms or data-intensive backend services.

• Strong software engineering skills in Python or a comparable backend language.

• Experience with data indexing, versioning, metadata, storage and access control.

• Experience evolving schemas or APIs while supporting existing users and historical data.

• Ability to lead technical design and implement substantial parts of the system.

• Ability to work with data or machine-learning users and translate their needs into reliable services.

Bonus qualifications and relevant technologies:

Experience in any of the following is beneficial but not required:

• Experience with LanceDB, Arrow, Parquet or comparable columnar and vector-data systems.

• Experience with dataset catalogs, provenance or reproducible machine-learning workflows.

• Experience applying domain models or ontologies to practical data relationships and schema evolution.

• Experience deploying data services on Kubernetes or in on-premises and disconnected environments.

Why join us:

The Data Platform is foundational to how Shield AI builds, evaluates, certifies, and improves autonomy. This role offers the opportunity to shape both a core internal platform and a reference architecture delivered into demanding customer environments.

Your work will determine how effectively engineers and agents can find trusted context, understand provenance, aggregate data from distributed systems, and turn operational experience into better intelligent systems. You will work at the intersection of data systems, AI infrastructure, autonomy, distributed computing, and defense while helping establish the architecture that supports the next generation of mission-critical AI development.