Freelancer profile translated to English.
Back to original languageAbout Gaëtan
I design and deployAI agentsandMCP serversconnected to your systems.
Inria Research Engineerwith 7 years of development experience: 2 years of freelance web development, 5 years of applied research.
What I do
- Custom AI Agents:LangGraph, MCP, RAG, function calling, etc. connected to your CRMs/ERPs/internal databases, via ChatGPT/Claude or custom client
- MCP Servers:your business tools accessible in natural language
- LLM Platforms:RAG on sensitive data, fine-tuning, monitoring. On-premise or European private cloud (healthcare, finance, R&D)
- AI for Science:Surrogate models, scientific ML, simulation acceleration, time series
Why me
- I deliver clean, typed, documented, and tested code. You get something you can review and evolve without me
- Strong experience in sensitive data processing, scientific rigor
- Active monitoring: I use the tools I offer daily
Stack
- Python (expert), C/C++, TypeScript
- LangChain/LangGraph, MCP, deepagents, Claude SDK
- OpenAI/Anthropic/Mistral, vLLM/ollama
- RAG (Pinecone, Qdrant, pgvector), PostgreSQL
- PyTorch, JAX, Hugging Face
- Docker, CI/CD, AWS/GCP/Azure/OVH, HPC (SLURM, OAR)
Collaboration
- Sophia Antipolis
- Short sprints, weekly demos, versioned code from day 1, documentation + tests included
- First 30-minute call free to scope your needs
French
Native or bilingual
English
Fluent
Remote only
Primarily works remotely
Experience
- InriaResearch EngineerMEDICALSeptember 2025 - Today (11 months)Design and deployment of aconversational AI assistantfor physicians, as part of a clinical research study. The agent allows natural language querying of heterogeneous patient data (measurements, exams, metadata) and generation of on-demand analyses.**Responsibilities**:
- Agent architecture: sandbox, advanced reasoning, and tools
- Secure connection to sensitive clinical data
- Design of analysis tools (extraction, aggregation, visualization) callable by the agent (MCP)
- Production deployment on infrastructure, provision of a web interface for physicians
- Iterative collection of user feedback and continuous improvements
GDPR confidentiality, reliability of responses on critical data, user-friendliness for non-technical users - InriaPhD - Cardiac Modeling, Numerical SimulationRESEARCHJanuary 2020 - March 2023 (3 years and 2 months)Nice, France
PhD Abstract
Heart failureis the final funnel of allcardiac diseases**, currently affecting 8% of the general population living in developed countries and is expected to reach 11% by 2030. Despite major improvements provided by optimization of medical therapy and prevention, this disease still has a high mortality rate and represents 1-2% of the total medical expenses. Among the different new treatments that became available in the last years, **Cardiac Resynchronization Therapy (CRT)has emerged as a very original technique used to correct mechanical abnormalities in these failed hearts, by direct stimulation of the myocardium in selected locations. This thesis describes the feasibility of a novel approach to use adigital twinin order to predict the response to CRT. The biophysical heart model ispersonalizedfrom routine patient data using fastartificial intelligence(AI)- based methods, and the output simulation results are compared to clinical data provided by a cardiac devices manufacturer.
Recommendations
Be the first to recommend Gaëtan
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Engineer's degree in Mechanics and Mathematical ModelingENSEIRB-MATMECA | Bordeaux University, France2019Engineer's degree, Mechanics and Mathematical Modeling