M

Applied Scientist - Research Engineer (Optimisation), EMEA

Mistral·Lausanne·04.11.2026

 jobs.ashbyhq.com

 
 
full_time80–100%
Required language
English professional
Job written in
English
Location
Lausanne
Work type
On-site
Type
Full-time

 

Mistral is a full‑stack AI company that builds frontier models, developer tools, applications and compute platforms for enterprises in finance, manufacturing, defence, healthcare and the public sector. The team is globally distributed, low‑ego and collaborative, and focuses on turning cutting‑edge research into practical AI systems that clients can run on their own infrastructure. The Applied Scientist, Research Engineer (Optimisation) will post‑train state‑of‑the‑art large language models and use them to formulate and solve optimisation problems such as linear programming, mixed‑integer programming, scheduling and routing. Daily work includes running pre‑ and post‑training on large GPU clusters, creating and curating datasets for optimisation modelling, building evaluators and benchmarks, and using solvers to generate reward signals for reinforcement learning. You will also develop tools and frameworks for data generation, model training, evaluation and deployment, and coordinate research projects with internal and client teams. The role requires fluent English communication and the ability to explain technical concepts to diverse audiences. You must be an expert in PyTorch or JAX and have experience training or fine‑tuning large language models for reasoning, code or mathematics. A strong background in operational research or mathematical optimisation is essential, with hands‑on use of solvers such as Gurobi, CPLEX or OR‑Tools and modelling languages like Pyomo or AMPL. You should write clean, high‑performance, fault‑tolerant Python code, work independently with minimal guidance, and demonstrate a track record of success through personal, professional or academic projects. A collaborative, low‑ego attitude and eagerness to learn are also required. Preferred but not mandatory attributes include having been an ex‑founder or builder, holding a PhD or Master's in operational research, applied mathematics or machine learning, and experience with reinforcement learning that uses verifiable rewards. Additional pluses are experience with LLM‑based optimisation modelling (e.g., NL4Opt), industry exposure to supply‑chain, logistics, energy or scheduling, work on large codebases or massive ML models, contributions to open‑source or industry repositories, publications in top venues, and a focus on improving code quality through typing, testing and CI pipelines. What the role asks for: - Fluent English communication - Expert in PyTorch or JAX - Experience training/fine‑tuning large language models - Strong background in operational research and optimisation - Hands‑on with solvers (Gurobi, CPLEX, OR‑Tools) - Hands‑on with modelling languages (Pyomo, AMPL) - Clean, high‑performance Python coding - Ability to work independently with little guidance - Low‑ego, collaborative attitude - Proven track record of project or academic success - (nice to have) Ex‑founder or builder - (nice to have) PhD or Master in relevant field - (nice to have) Reinforcement learning with verifiable rewards - (nice to have) LLM‑based optimisation modelling experience - (nice to have) Industry experience in supply chain, logistics, energy or scheduling - (nice to have) Experience with large codebases or large ML models at scale - (nice to have) Contributions to large open‑source or industry codebase - (nice to have) Publications in top academic venues - (nice to have) Focus on code quality: typing, tests, CI pipelines

 

 

 

 

 

Kostenlos

Scientist: Lebenslauf-Vorlage

Bewirb dich mit einem Lebenslauf im Schweizer Aufbau — mit Beispieltext und den Anforderungen, die Scientist-Inserate am häufigsten nennen.

Lebenslauf-Vorlage Schweiz ansehen

Scientist: was der Markt gerade verlangt

Scientist-Stellen gehören zu den regelmässig ausgeschriebenen Berufen auf SwissJobs.app.

Weitersuchen

Ähnliche Jobs per E-Mail erhalten