Applied Scientist / Research Engineer - Edge Devices and Quantization - EMEA

Mistral AI • Paris, TX

64d

About The Position

Mistral AI is seeking Applied Scientists and Research Engineers focused on model efficiency and edge deployment. You will research and build ultra‑efficient models and toolchains for on‑device inference across CPUs, GPUs, NPUs, and specialized accelerators. Your work will enable Mistral models to run privately, reliably, and fast on mobile, desktop, and embedded devices.

Requirements

• You are fluent in English, and have excellent communication skills. You are at ease explaining complex technical concepts to both technical and non-technical audiences.
• You're not afraid of contributing to a big codebase and can find yourself around independently with little guidance.
• You've a deep understanding of quantization trade‑offs, hardware constraints, and compiler stacks
• You're expert with PyTorch or JAX; strong C++/CUDA or low‑level performance skills a plus; production‑grade Python.
• You don't need roadmaps: you just do. You don't need a manager: you just ship.
• Low-ego, collaborative and eager to learn.
• You have a track record of success through personal projects, professional projects or in academia.

Nice To Haves

• Hold a PhD / master in a relevant field (e.g., Mathematics, Physics, Machine Learning), but if you're an exceptional candidate from a different background, you should apply.
• Have contributed to a large codebase used by many (open source or in the industry).
• Have a track record of publications in top academic journals or conferences.
• Contributions to open‑source inference/compilers stacks.
• Love improving existing code by fixing typing issues, adding tests and improving CI pipelines.
• Have experience optimizing inference on edge devices

Responsibilities

• Run pre-training, post-training and deploy state of the art models on clusters with thousands of GPUs. You don't panic when you see OOM errors or when NCCL feels like not wanting to talk.
• Design and evaluate quantization, pruning, distillation, and sparsity methods for LLMs and multimodal models.
• Build deployment stacks, optimize kernels and memory layouts.
• Run large‑scale experiments to balance accuracy, latency, throughput, and power under tight memory constraints; profile and fix bandwidth/compute bottlenecks.
• Develop tooling for calibration data generation, mixed‑precision training, quant‑aware finetuning, structured/unstructured sparsity, and compilation passes.
• Manage research projects and communications with client research teams.

Benefits

• Competitive cash salary and equity
• Food : Daily lunch vouchers
• Sport : Monthly contribution to a Gympass subscription
• Transportation : Monthly contribution to a mobility pass
• ️ Health : Full health insurance for you and your family
• Parental : Generous parental leave policy
• Visa sponsorship
• Insurance
• Transportation: Reimburse office parking charges, or 90GBP/month for public transport
• Sport: 90GBP/month reimbursement for gym membership
• Meal voucher: £200 monthly allowance for its meals
• Pension plan: SmartPension (percentages are 5% Employee & 3% Employer)

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Industry: Publishing Industries
• Number of Employees: 251-500 employees