Research Engineer, Data Infrastructure
Mistral AI2 months ago
Paris, France +3 moreMid Level
Responsibilities
- Build and scale distributed compute and storage systems for large-scale model training and fine-tuning.
- Architect and maintain multi-cluster orchestration layers across regions and hardware environments.
- Design modern storage systems and formats for fine-tuning datasets with anticipated exabyte-scale growth.
- Develop the internal training platform across Kubernetes- and SLURM-based environments.
- Implement metadata and data/model lineage systems.
- Manage cloud-native deployments, production pipelines, reliability, secure data access, and on-call support for critical training jobs.
Requirements
- 4+ years of experience in data infrastructure, MLOps, or infrastructure engineering.
- Experience or strong interest in foundational compute and storage platforms.
- Proficiency in Python and interest in modern columnar storage standards.
- Experience with Kubernetes-native tooling and debugging large-scale distributed systems in multi-cluster environments.
- Ability to build and operate scalable, reliable, and secure systems.
Benefits
- Benefits may include healthcare coverage, parental leave, retirement plans, relocation support, wellness programs, and meal and transportation allowances, varying by country.
- Hybrid work is available in Paris, London, Warsaw, or Zurich, with remote options in France, the UK, Poland, and Switzerland.
- Remote hires must attend the local hub during the first onboarding week and at least three days per month; travel and accommodation for onboarding are covered.
Tech Stack
Categories
Data EngineeringDevOps
About Mistral AI
Mistral AI builds foundation language models and full‑stack AI solutions for enterprises, offering APIs, on‑prem deployment, developer tools, and applications. The privately held company, founded in 2023 and headquartered in Paris, partners with organizations in finance, manufacturing, defense, healthcare, and the public sector to co‑create customized systems. It releases open‑source models alongside commercial offerings and makes its models available through major cloud marketplaces.