Data Engineer: Build Clean, Versioned ML Datasets
Summary
Build and maintain clean, versioned datasets for ML training by ingesting and curating terabytes of raw data, ensuring reproducibility and collaboration across teams.
Harmattan AI is seeking a Data Engineer to own the data layer for deep-learning pipelines in Paris, Lausanne, or Zurich. You will transform terabytes of raw, unstructured data into clean, versioned, training-ready datasets so modelers can focus on model design and training.
The role emphasizes ingestion, curation, labeling workflows, and reproducible data lineage, with collaboration across acquisition, annotation, and product teams.