The Role
Own Doctolib’s self-hosted LLM serving platform, deploying and scaling open-weight models across GPU infrastructure. Responsibilities include serving-stack design, model registry and promotion, inference optimization, benchmarking, reliability, autoscaling, disaster recovery, observability, cost reporting, and security compliance. The role requires production experience with LLM serving engines, GPU workloads, Kubernetes or Ray, performance engineering, and on-call operations.
Summary Generated by Built In
Set a new pulse for healthcare!
We are looking for a Senior Inference Engineer to join the ML Platform (TIGER) team in AI & Machine Learning.
Your mission will be to power Doctolib's AI-driven healthcare products — from Consultation Assistant to Phone Assistant and Health Companion — by owning the self-hosted LLM serving platform that keeps the most sensitive patient data within infrastructure Doctolib controls. You will build and operate the shared foundations that every AI feature team relies on: deploying, tuning, and scaling open-weight models across our GPU fleet, contributing directly to faster, safer, and more cost-efficient AI for care teams and patients across Europe.
Working in the tech team at Doctolib means building innovative products and features to improve the daily lives of care teams and patients.
What you'll doYour responsibilities include but are not limited to:
- Build the shared LLM serving foundations with MLOps engineers — define the serving stack, GPU capacity planning, and deployment automation for open-weight models (Gemma, Mistral, Llama, Qwen, and their fine-tuned variants) on Ray Serve / Anyscale, while shaping the evolution toward Kubernetes-based orchestration
- Own the model registry and promotion path — establish clear processes for versioning, validating, and promoting models into production, benchmark performance and output quality under realistic workloads, and maintain the curated model catalog together with the Agents Platform team
- Optimize inference performance and efficiency — operate and tune serving engines (vLLM, Triton Inference Server, TensorRT-LLM) to reduce time-to-first-token, inter-token latency, and cost per token through batching, caching, quantization, and parallelism strategies across our GPU fleet (L4 to H200)
- Own production reliability and disaster recovery — define SLOs, run load tests before launches, implement safe rollouts, rollbacks, and autoscaling, and participate in on-call and incident response
- Make serving performance and costs measurable — build Datadog dashboards, alerting, and cost reporting to track latency, GPU utilization, and inference spend, and identify opportunities to improve resource efficiency
- Keep the inference stack secure, compliant, and evolving — work with Security and Privacy on network isolation, data residency, and auditability, while continuously assessing new models, serving engines (such as SGLang), and GPU generations
Before you read on: if you don't have the exact profile described below, but you feel this job description matches your skill set, we still encourage you to apply.
You'll be a great fit if you:
- Have 5+ years of software or computer engineering experience, including hands-on production experience with at least one LLM serving engine (vLLM, SGLang, TensorRT-LLM, TGI, or Triton) — ideally having chosen, introduced, tuned, or scaled one in production, not just run it with defaults
- Are proficient in Python and Terraform, with a solid understanding of transformer inference mechanics: prefill vs. decode, KV cache, batching strategies, and why a model is compute- or memory-bound on a given GPU
- Have strong experience running GPU workloads on Kubernetes or a managed Ray / Anyscale platform, and a performance engineering mindset — you design reproducible benchmarks, read profiler output, and back decisions with numbers
- Are comfortable owning production reliability: observability, incident response, and on-call are a natural part of the role
- Have excellent communication and collaboration skills, high agency, a pragmatic build-vs-buy approach, and are comfortable working in a scope that is still taking shape
- Are fluent in English
It would be fantastic if you:
- Have experience with distributed serving frameworks (KServe, llm-d, NVIDIA Dynamo), including disaggregated prefill/decode or KV-cache-aware routing, or with serving Mixture-of-Experts, multimodal models, or many LoRA adapters on shared base models
- Have GPU kernel experience or deep profiling skills (CUDA, Triton, Nsight), or have contributed to open-source inference projects
- Have experience in regulated environments (healthcare, finance, public sector) and with frameworks such as GDPR, HDS, ISO 27001, or SOC 2
- Programming languages: Python, plus comfort reading/debugging a systems language (C++, Rust, or Go)
- Cloud providers: GCP / AWS / Azure
- Serving orchestration: Ray Serve / Anyscale (GCP), with Kubernetes (GKE) + GPU scheduling as an emerging parallel target
- LLM serving engines (in production): vLLM, Triton Inference Server, TensorRT-LLM (also evaluating SGLang and other emerging engines)
- Models: Self-hosted open-weight LLMs (Gemma, Mistral, Llama, Qwen), fine-tuned in-house
- Experiment tracking & model registry: MLflow
- GenAI gateway: LiteLLM
- LLM evaluation & observability: Braintrust
- GPUs: NVIDIA L4, A10G, L40S, RTX 6000 Pro, A100 (40GB/80GB), H200
- Container / Infra as code: Docker, Terraform, Helm, Kustomize
- Observability: Datadog
- Data warehouse / storage: GCP BigQuery / GCS / AWS S3
- And any other tool you deem relevant!
- Our solutions are built on a single fully cloud-native platform that supports web and mobile app interfaces, multiple languages, and is adapted to country and healthcare specialty requirements.
- Our stack is composed of Rails, TypeScript, Java, Python, Kotlin, Swift, and React Native.
- We leverage AI ethically across our products to empower patients and health professionals. Discover our AI vision here.
Want to learn more about our tech culture and environment? Visit the Doctolib Tech site.
What we offer- Free comprehensive health insurance (basic package) for you and your children
- 25 days of paid vacation per year, plus up to 14 days of RTT
- Free mental health and coaching services through our partner Moka.care
- Work from abroad for up to 10 days per year thanks to our flexibility days policy
- Lunch vouchers (Swile card) worth €8.50 per working day, with €4.50 covered by Doctolib
- A subsidy from the work council to refund part of the membership to a sport club or a creative class
- 50% reimbursement of your public transport subscription
- ParentCare Program: Enjoy full salary coverage (100%) during your first month of birth leave and 75% during the second, covered by Doctolib
- Enrollment in Doctolib's long-term employee value sharing plan called DoctoGrowth
- For caregivers and workers with disabilities, a package including an adaptation of the remote policy, extra days off for medical reasons, and psychological support
- Relocation support in case of international mobility
- Access to the best AI tools for coding, development and dedicated training
- HR interview by phone (45 minutes)
- Hiring manager interview (1 hour)
- Technical case study (1 hour)
- Behavioral interview (1 hour 30 minutes)
- At least one reference check
We want your experience to be clear, respectful, and transparent. Learn more about our hiring process on our candidate experience page.
Job details- Permanent position
- Tech stack: Python, vLLM, TensorRT-LLM, Triton, Ray Serve
- Full-time
- Paris, France
- Hybrid work setup (up to 2 remote days per week)
- Start date: as soon as possible
At Doctolib, we are committed to improving access to healthcare for everyone. This translates into our recruitment process. We evaluate candidates based solely on qualifications and motivation, without any form of discrimination.
The more diverse ideas are heard, the more our product will truly improve healthcare for all. You are welcome to apply to Doctolib, regardless of your gender, religion, age, sexual orientation, ethnicity, or disability.
To ensure equal opportunities, we invite you to exclude personal information (e.g., pictures, age) from your applications. If you require any accommodation, please let us know for support during the hiring process.
Join us in building the healthcare we all dream of!
Your data privacyAll information provided is processed by Doctolib for application management. For data processing details, click here: France. Please contact hr.dataprivacy(at)doctolib.com for inquiries or to exercise your rights.
Skills Required
- 5+ years of software or computer engineering experience
- Hands-on production experience with at least one LLM serving engine such as vLLM, SGLang, TensorRT-LLM, TGI, or Triton
- Proficiency in Python and Terraform
- Understanding of transformer inference mechanics, including prefill, decode, KV cache, batching, and GPU compute or memory constraints
- Experience running GPU workloads on Kubernetes or a managed Ray/Anyscale platform
- Performance engineering experience, including reproducible benchmarks and profiler analysis
- Production reliability experience with observability, incident response, and on-call operations
- Excellent communication and collaboration skills
- High agency and pragmatic build-versus-buy decision-making
- Fluency in English
- Experience with distributed serving frameworks such as KServe, llm-d, or NVIDIA Dynamo
- Experience with disaggregated prefill/decode, KV-cache-aware routing, Mixture-of-Experts, multimodal models, or LoRA adapters
- GPU kernel experience or deep profiling skills with CUDA, Triton, or Nsight
- Contributions to open-source inference projects
- Experience in regulated environments such as healthcare, finance, or the public sector
- Knowledge of GDPR, HDS, ISO 27001, or SOC 2
Am I A Good Fit?
Get Personalized Job Insights.
Our AI-powered fit analysis compares your resume with a job listing so you know if your skills & experience align.
Success! Refresh the page to see how your skills align with this role.
The Company
What We Do
Since Doctolib's creation in 2013, we have had one purpose: strive for a healthier world. 1. We aim to improve the daily lives of care teams by providing them with a new generation of technologies and services. 2. We aim to improve health for all, by offering a fast and frictionless journey for all care episodes, creating new ways for people to receive care and empowering them to become actors of their health. At Doctolib, we are honored to work in the healthcare field and we believe that innovation in healthcare should be handled differently. We apply 4 guiding principles in everything we do: 1. We create helpful solutions for care teams and people. 2. We serve everyone equally and create well-designed and accessible technologies. 3. We team up with our users to strive for a healthier world and act as one team. 4. We protect our users' privacy. It’s their health, their data. To achieve our purpose, we are assembling a team dedicated to improving healthcare, with a human-centric approach and an entrepreneurial mindset. www.doctolib.com
