Member of Engineering (Pre-training / Data Engineering)

📋 Description

• Build and maintain high-performance pipelines for trillions of tokens.
• Deliver diverse, high-quality datasets for pre-training foundation models.
• Collaborate with Pretraining, Posttraining, Evals & Product to ensure model quality.

🎯 Requirements

• Strong background in production-grade, distributed data systems for ML
• Orchestration: Slurm, Airflow, or Dagster
• Observability & Reliability: CI/CD, Grafana, Prometheus
• Infra: Git, Docker, Kubernetes, cloud managed services
• Batch inference (vLLM) and GPU cluster optimization
• Python expert; clean code; Polars/Dask/PySpark proficiency

🎁 Benefits

• Fully remote work & flexible hours
• 37 days/year vacation & holidays
• Health insurance allowance for you & dependents
• Company-provided equipment
• Well-being and home office allowances
• Frequent team get togethers