Inference Systems Engineer – Scalable LLM Serving

Lindus Health Limited; Warszawa; Ibrido; PLN 180.000 - 300.000; Tempo pieno

Descrizione del lavoro

Mistral is seeking a senior engineer to own the core of its inference stack, including the engine and orchestration, with a focus on performance and scale. You will help power frontier model training and ensure elastic capacity across production workloads.

The role spans production LLM serving, engine development, and capacity engineering in a hybrid setup. Candidates should have deep ML/LLM infra experience and strong systems expertise.

Competenze

• Experience building and running ML/LLM services atScale with latency targets.
• Hands-on with inference engines such as vLLM, sgLang, TensorRT-LLM, or similar.
• Strong grasp of inference internals and distributed serving architectures.

Mansioni

• Implement and optimize the core inference stack and orchestration.
• Own release processes for serving stack with automated gates.
• Drive improvements upstream when appropriate.

Conoscenze

Python

PyTorch

Kubernetes

CUDA/NCCL

InfiniBand/RDMA

vLLM/sglang

Go/TensorRT-LLM

Strumenti

NCCL

Containers

Networking

Descrizione del lavoro

Mistral is seeking a senior engineer to own the core of its inference stack, including the engine and orchestration, with a focus on performance and scale. You will help power frontier model training and ensure elastic capacity across production workloads.

The role spans production LLM serving, engine development, and capacity engineering in a hybrid setup. Candidates should have deep ML/LLM infra experience and strong systems expertise.