Data Engineer, Retrieval & Training Pipelines

Atlas runs on operational data: events, inventories, movements, sensor feeds, and the records that operators and institutions generate every day. This role owns the systems that find that data, retrieve it correctly, prepare it for training and inference, and keep it flowing reliably as the platform grows.

• Build and operate the retrieval layer: how Atlas finds and pulls operational data from diverse sources, with correct provenance and audit trail.
• Design and maintain the training data pipelines: ingestion, cleaning, feature engineering, and versioning for machine learning workloads.
• Own the schema and lineage of the data used for training. Models are only as trustworthy as the data behind them.
• Work with intelligence engineers on feature stores, offline and online consistency, and training-serving skew.
• Build tooling that makes data exploration and analysis fast for analysts and scientists.

• Strong data engineering background in production environments.
• Experience with analytical pipelines and ETL in production (dbt, Airflow, Spark, or equivalent).
• Strong SQL and comfort with at least one systems language (Python, Go, or similar).
• Attention to provenance and data quality. These are not afterthoughts for us.

These are not required. They are the kind of background that tends to do well in this role.

• Experience building retrieval or feature pipelines for machine learning systems.
• Experience with event streaming or event-sourced architectures.
• Geospatial data experience.