Data Infrastructure Engineer — Scalable AI Platform

Mistral; Paris; Hybrid; EUR 144,000 - 227,000; Full time

Benefits offered by this job

Equity

Health insurance

Lunch vouchers

Gympass

Mobility subsidy

Parental leave

Visa sponsorship

Job summary

Mistral AI is building a high-performance data infrastructure for frontier model training and fine-tuning. You will design and scale massive compute fleets and storage systems, enabling secure, governed data access for MLOps and research. The role covers on-premise and cloud-native Kubernetes environments.

Location options include Paris or London (hybrid) or remote EU/UK with one hub day per month. You will own end-to-end pipelines, migrate from legacy schedulers, and contribute to on-call

Qualifications

• 4+ years of experience in Data Infrastructure, MLOps, or Infrastructure Engineering.
• Experience or interest in supporting foundational compute and storage platforms.
• Proficient in Python and solving data lake problems with modern storage standards.
• Well-versed in Kubernetes-native tooling; comfortable debugging large-scale distributed systems across multi-cluster environments.
• Ability to build scalable, reliable, and secure systems from the ground up.
• Comfortable with ambiguity in a rapid-growth AI environment.

Responsibilities

• Build & Scale: operate massive distributed compute and storage systems.
• Global Orchestration: maintain multi-cluster orchestration layers across diverse hardware and regions.
• Design Future-Proof Storage: transition to modern storage formats for exabyte-scale datasets.
• Platform Engineering: develop internal training platform for model training and fine-tuning on Kubernetes and SLURM.
• Metadata & Lineage: provide visibility and lineage for data and model pipelines.
• Operational Excellence: manage cloud-native deployments for scalable data platforms.

Skills

Data Infrastructure

MLOps

Infrastructure Engineering

Python

Kubernetes

Multi-cluster orchestration

Security & governance

Tools

Kubernetes

SLURM

Job description

Mistral AI is building a high-performance data infrastructure for frontier model training and fine-tuning. You will design and scale massive compute fleets and storage systems, enabling secure, governed data access for MLOps and research. The role covers on-premise and cloud-native Kubernetes environments.

Location options include Paris or London (hybrid) or remote EU/UK with one hub day per month. You will own end-to-end pipelines, migrate from legacy schedulers, and contribute to on-call