Research Engineer: End-to-End AI Training & Deployment

Mistral; Lausanne; On-site; CHF 120,000 - 180,000; Full time

Job summary

Mistral is seeking an experienced Research Engineer to build end-to-end training and deployment workflows for Forge. You will bridge research experiments with production constraints, collaborating with researchers and infrastructure teams to ship reliable solutions.

The role emphasizes strong Python software engineering, PyTorch/JAX experience, and deep systems knowledge to support post-training workflows and scalable pipelines for client projects.

Qualifications

• Strong Python engineering skills in large codebases.
• Hands-on experience with PyTorch or JAX.
• Strong systems and infrastructure fundamentals.
• Experience with LLM training or post-training workflows.
• Excellent debugging skills in distributed environments.
• Clear communication with technical and non‑technical stakeholders.

Responsibilities

• Build and improve post-training and evaluation workflows.
• Develop tools and pipelines for data generation, training, evaluation, and deployment.
• Harden large-scale ML systems: distributed training, scheduling, observability.
• Improve codebase with APIs, tests, and maintainable abstractions.
• Advance RL training stack for frontier-model scale.
• Ensure Forge deployment is adaptable across diverse clients.

Skills

Python

Codebase ownership

PyTorch/JAX

Systems & Infra

LLM training

Debugging

Communication

Autonomy

Tools

PyTorch

JAX

SLURM

Kubernetes

Ray

Job description

Mistral is seeking an experienced Research Engineer to build end-to-end training and deployment workflows for Forge. You will bridge research experiments with production constraints, collaborating with researchers and infrastructure teams to ship reliable solutions.

The role emphasizes strong Python software engineering, PyTorch/JAX experience, and deep systems knowledge to support post-training workflows and scalable pipelines for client projects.