Applied Scientist - Formal Methods, EMEA
Paris • Full-time
Posted 1mo ago
Job Location
Paris
Tech Stack
Remote Work Policy
On-site
Employment Type
Full-time
Categories
AI Research Engineer
About the job
Mistral AI is seeking Applied Scientists and Research Engineers focused on formal methods to design, train, and deploy state-of-the-art large language models and apply them to diverse use cases. You will work cross-functionally with internal and external science, engineering, and product teams to deliver high-impact AI solutions. The role involves running pre-training and post-training of models on large clusters, generating and curating datasets for theorem proving and software verification, and building evaluators and benchmarks for formal reasoning. You will also develop tools and frameworks to facilitate data generation, model training, evaluation, and deployment, while collaborating with teams to tackle complex use cases and managing research projects and client communications.
Responsibilities
- Run pre-training, post-training, and deployment of state-of-the-art models on large clusters.
- Generate and curate datasets for theorem proving and software verification.
- Build evaluators and benchmarks for formal reasoning.
- Develop tools and frameworks for data generation, model training, evaluation, and deployment.
- Collaborate with cross-functional teams to address complex use cases.
- Manage research projects and client research team communications.
Requirements
- Fluent in English with excellent communication skills, able to explain complex technical concepts to diverse audiences.
- Expertise in PyTorch or JAX.
- Proficiency in writing clean, readable, high-performance, fault-tolerant Python code.
- Ability to navigate uncertainty and work independently with little guidance.
- Experience with autoformalization, proof automation, or verified code generation (nice to have).
- Experience with reinforcement learning using verifiable rewards, ideally from Lean or other proof assistants (nice to have).
- Contributions to a large open-source or industry codebase (nice to have).
- Publications in top academic journals or conferences (nice to have).
- Experience improving existing code, including typing, tests, and CI pipelines (nice to have).
Benefits
- Healthcare coverage
- Parental leave
- Retirement plans
- Relocation support
- Wellness programs
- Meal and transportation allowances
