Research Engineer GPU Performance
Mistral AIVisit Mistral AI website
Paris-based AI company building frontier models, AI applications, developer tools, and compute infrastructure for enterprise and public-sector deployments.
Paris, France
About Mistral AI
Mistral AI develops open-weight and commercial language models and provides a full-stack AI platform spanning agents, application development, custom-model training, APIs, and AI cloud infrastructure.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will write low-level GPU code and optimize model architectures for efficient large-scale training and inference. You will integrate performance work into MLOps systems, profile and benchmark workloads across GPU clusters, and model LLM execution to estimate hardware use and serving performance.
Requirements
- Expertise writing custom low-level kernels
- Expertise with distributed GPU cluster infrastructure
- Knowledge of generative AI
- Knowledge of fine-tuning and language model applications
- Software design knowledge, including testing, code review, and CI/CD
Responsibilities
- Develop low-level GPU code, custom kernels, memory layouts, and communication optimizations
- Rethink generative-model architecture for efficient inference
- Integrate optimized code into the MLOps framework
- Profile and benchmark large-scale training and inference
- Model LLM execution across GPUs and estimate serving performance