Research Engineer, Latency & Throughput
Mistral AIParis, LondonOn-Site
Needs a free Superjobs account. You'll finish on Mistral AI's Ashby page.
Skill score
Summary
The Research Engineer on the Latency & Throughput team researches and prototypes LLM inference techniques such as speculative decoding, quantization, and KV compression, runs experiments, and helps ship validated methods into the serving stack. It suits a science-minded engineer with genuine research depth, strong Python and PyTorch skills, and comfort with ambiguity and negative results.
- Function
- Development
- Estimated salary
- Est. $46k–69k/yr
- Estimated for a mid-level role, ~3 yrs experience (typical for the level), France. Not published by the employer.
- Work mode and location
- On-Site, Paris, London
- Employment type
- Full-time
Job description
Overview
As a Research Engineer on the Latency & Throughput team, you will push the frontier of LLM inference by researching and prototyping techniques that reduce latency and increase throughput — from speculative decoding and quantization to KV compression and new attention variants. The team sits at the intersection of research and production: you will take ideas from recent literature, form hypotheses, run experiments, and ship the techniques that prove out into our serving stack.
About the team
This role sits between our inference engineers, who own the production serving stack, and our research scientists, who focus on paper-level contributions. You will be more science-focused than a pure engineer: you are expected to generate new ideas, keep up with the state of the art, and implement proof-of-concepts that bridge the gap between academic results and production impact. Some techniques you explore will be lossy or require integration into training — retraining models, tuning architectures, or training auxiliary models to improve inference.
You will work closely with the Inference Foundation and Kernels teams, who turn your validated techniques into production recipes, and with pretraining and RL teams when your work touches model training. Success is measured by demonstrated latency or throughput gains, first at small scale, then adopted in production.
Roles and responsibilities
- Research and prototype inference techniques — speculative decoding, quantization, KV compression, watermarking, and new model architectures — by reading recent literature and forming testable hypotheses.
- Design and run experiments to validate or refute ideas, iterating quickly on small-scale setups before proposing production integration. Demonstrate clear latency or throughput gains without unacceptable quality loss.
- Collaborate with the Inference Foundation team to integrate validated techniques into the vLLM-based serving stack.
- Partner with pretraining and RL teams when techniques require co-training, retraining, or training auxiliary models to improve inference.
Build the right tooling and validation platforms to facilitate AI-driven inference research.
Skills required
Top skills
- LLM inference
- LLM training techniques
- Python
- PyTorch
All skills (8)
- Experiment design
- Research experience
- CUDA
- JAX
- vLLM
- Model quantization
- Speculative decoding
- Serving systems
Qualifications
- A strong ML background with solid knowledge of LLM training techniques — architecture, tokenization, fine-tuning, and evaluation.
- Hands-on experience with Python and PyTorch; you write clean, experimental code and iterate fast. JAX is acceptable with strong Python proficiency. C++ and Rust are not required, CUDA and GPU knowledge is expected but hands-on experience is not required.
- Research experience of some form: a thesis, publications, a dedicated research project, or efficiency-oriented work in a lab or industry research setting. A PhD is not required, but we are looking for genuine research depth.
- A track record of conducting experiments: forming hypotheses, designing controlled comparisons, and drawing clear conclusions from results.
- Familiarity with the state of the art in LLM inference — you know what has been tried, what works, and where the open problems are.
- Self-starter, low-ego, and collaborative. You are comfortable with ambiguity, negative results, and are eager to take matters into your own hands.
Nice to have
- Inference or serving experience is a plus, not a requirement. Prior work with vLLM, production LLM deployments, or serving systems will help you ramp up faster.
Other details
This role is based in our Paris or London offices (hybrid). We will prioritize candidates who reside in the Paris area or are open to relocating. We strongly believe in the value of in-person collaboration to foster strong relationships and seamless communication within our team. For senior profiles, we will also consider remote candidates based in France. In that case, we ask all new hires to visit our Paris office:
- For the first week of their onboarding (accommodation and travel covered)
- Then at least 3 days per month
Benefits
We offer a comprehensive benefits package designed to support your well-being, growth, and work-life balance. Benefits vary by country and may include healthcare coverage, parental leave, retirement plans, relocation support, wellness programs, meal and transportation allowances, and other location-specific perks.
For the most up-to-date details on benefits available in your location, please refer to our Benefits page.
Company info
About Mistral AI
Mistral provides full-stack AI solutions: from frontier models to developer tools, applications, and compute. We partner with enterprises tackling the hardest problems across high-stakes industries like finance, manufacturing, defense, healthcare, and the public sector, co-creating customized AI systems that they can run on their terms.