Senior LLM Inference Engineer — GPU‑Driven Serving

Doctolib; Levallois-Perret; Hybrid; EUR 110,000 - 150,000; Full time

Benefits offered by this job

Health insurance

Lunch vouchers

Remote work flexibility

Transport subsidy

Doctolib employee benefits

Job summary

Doctolib is seeking a Senior Inference Engineer to lead the self-hosted LLM serving platform within the AI & Machine Learning team. You will scale open-weight models on GPU fleets, deploy and tune serving engines, and own the model registry while improving safety, cost, and performance for healthcare AI features across Europe.

You will work in a hybrid Paris-based role with a strong emphasis on MLOps, Kuberenetes orchestration, and collaboration across security and privacy teams to keep patient

Qualifications

• 5+ years of software or computer engineering with production LLM serving experience.
• Proficient in Python and Terraform and understanding of transformer inference.
• Strong experience running GPU workloads on Kubernetes or Ray/Anyscale.
• Comfortable owning production reliability, observability, and on-call duties.
• Excellent communication, collaboration, and English fluency.

Responsibilities

• Build shared LLM serving foundations with MLOps engineers and define the serving stack.
• Own the model registry, versioning, validation, and production promotion workflows.
• Optimize inference performance and cost with batching, caching, and quantization.
• Ensure reliability with SLOs, load testing, safe rollouts, and auto-scaling.
• Build dashboards and cost reporting to monitor latency and GPU usage.
• Maintain security, data residency, and auditability in collaboration with Security.

Skills

Python

Terraform

LLM serving

English fluency

Tools

Ray Serve

Kubernetes

Docker

vLLM

TensorRT-LLM

Triton

Job description

Doctolib is seeking a Senior Inference Engineer to lead the self-hosted LLM serving platform within the AI & Machine Learning team. You will scale open-weight models on GPU fleets, deploy and tune serving engines, and own the model registry while improving safety, cost, and performance for healthcare AI features across Europe.

You will work in a hybrid Paris-based role with a strong emphasis on MLOps, Kuberenetes orchestration, and collaboration across security and privacy teams to keep patient