Data Engineer (Detect & Track Distillation)
Harmattan AI3 months ago
Paris, France or Zürich, SwitzerlandSenior
Responsibilities
- Ingest, decode, and store raw video and sensor data from field logs in efficient, controlled formats.
- Align multimodal data streams temporally and spatially for model training.
- Curate and transform ingested and public datasets through parsing, filtering, de-duplication, and revision.
- Define data collection and labeling requirements and own dataset-construction and labeling tooling.
- Build task-specific training and evaluation datasets with acquisition, annotation, and product teams.
- Version datasets and maintain lineage to keep training runs reproducible.
- Manage efficient training-ready storage formats and storage tiering as datasets scale.
- Deliver documented datasets and data-loading tools optimized for model training performance.
Requirements
- A degree in a STEM field or equivalent practical experience.
- Experience building and maintaining large-scale pipelines for unstructured video, image, or sensor data covering ingestion, decoding, storage, curation, and versioning.
- Strong Python and data engineering skills.
- Experience optimizing data loaders for common training frameworks such as PyTorch.
- Bonus experience with multimodal sensor data, ML labeling or dataset construction, dataset versioning tools, or distributed data processing tools.
- Systematic, quality-minded, pragmatic, service-oriented approach to automating messy data workflows.
- Commitment to Harmattan AI’s mission of providing a defensive edge to allied nations through ethical, high-impact technology.
Benefits
- The role is based in Paris, Lausanne, or Zurich.
- The position offers early-stage influence over how field and public data is processed for deep-learning pipelines.
About Harmattan AI
Harmattan AI builds autonomous defense systems—air and multi-domain robots and the software that coordinates them—for military and government customers. Its products span autonomy stacks for high-speed flight, guidance and control, computer vision, and radar-enabled sensing, delivered as deployable platforms and integrated software. Founded in 2024 and headquartered in Paris, the privately held company operates as a defense prime serving allied forces.