Find your next role
Discover roles across ambitious teams
5,412 jobs
- More jobs
Mistral
Research Engineer, Latency & Throughput
Paris, France · Hybrid · Full time · Mid level
About the role
Research and prototype techniques to reduce latency and increase throughput in LLM inference. Collaborate with teams to integrate validated methods into production serving stacks.
What you’ll do
• Research and prototype inference techniques. • Design and run experiments to validate ideas. • Collaborate with Inference Foundation team. • Partner with pretraining and RL teams. • Build tooling and validation platforms.
What you’ll bring
• Strong ML background with LLM training knowledge. • Experience with Python and PyTorch. • Research experience or equivalent depth. • Track record of conducting experiments. • Familiarity with state of the art in LLM inference.
Life at Mistral
Comprehensive benefits including healthcare, parental leave, retirement plans, relocation support, and wellness programs. Hybrid work model with emphasis on in-person collaboration.
About Mistral
Mistral provides full-stack AI solutions partnering with enterprises across industries to co-create customized AI systems.