Inference Systems Engineer: Scale & Performance

Mistral; Paris; Hybrid; EUR 120,000 - 180,000; Full time

Benefits offered by this job

Healthcare coverage

Relocation support

Wellness programs

Job summary

Mistral is seeking an experienced engineer to own and evolve the inference foundation that powers production LLM serving and frontier model training. You will optimize the core stack, manage releases, and push upstream improvements to high-performance serving systems.

The role blends engine, platform development, and capacity engineering in a hybrid setup. You will work across CUDA, NCCL, and distributed architectures to deliver low-latency, scalable inference.

Qualifications

• Experience building and running ML/LLM services at scale with latency and availability targets.
• Hands-on experience with inference engines such as vLLM, SGLang, TensorRT-LLM, or others.
• Solid grasp of inference internals: prefill vs. decode, KV-cache behavior, batching, scheduling, speculative decoding, parallelism strategies.
• Familiarity with distributed and disaggregated serving architectures.
• Comfortable debugging across the full stack — CUDA/NCCL, kernels, containers, networking, storage.
• Python for systems tooling and backend services; PyTorch.
• Kubernetes for running infrastructure at scale.
• GPU and networking fundamentals: CUDA runtime, NCCL, InfiniBand/RDMA

Responsibilities

• Develop and fix the core of the inference stack — engine and orchestrator — including feature selection, configuration, and tuning for maximum performance at scale.
• Own the release process for the serving stack: validated, regression-free releases through automated performance gates and progressive rollout.
• Drive improvements and fixes upstream when the open-source engine is the right place for them.
• Optimize serving efficiency across the fleet — driving down pod startup time and cache-related regressions at scale.
• Optimize and maintain the optimal serving topology — overlap communication and computation, ensure placement and routing.
• Build the serving infrastructure that powers RL and post-training for frontier models.
• Optimize inference performance across the full spectrum of workloads.

Skills

ML/LLM services

Inference engines

CUDA/NCCL

Kubernetes

Python

PyTorch

GPU performance

Tools

vLLM

SGLang

TensorRT-LLM

Job description

Mistral is seeking an experienced engineer to own and evolve the inference foundation that powers production LLM serving and frontier model training. You will optimize the core stack, manage releases, and push upstream improvements to high-performance serving systems.

The role blends engine, platform development, and capacity engineering in a hybrid setup. You will work across CUDA, NCCL, and distributed architectures to deliver low-latency, scalable inference.