Freelancer profile translated to English.

Description

Full Stack Developer (Python, TypeScript, React) specializing in the design of industrial AI systems. I build end-to-end LLM solutions: multi-agent systems, RAG pipelines, business workflow automation (Power Automate, Microsoft 365), guardrail integration, and agent evaluation. 4 years of experience in demanding environments (medical devices, industry) have trained me in the constraints of reliability, traceability, and data security. Essential for moving from POC to production. In parallel, I support French SMEs and mid-sized companies on their AI projects: opportunity audits, LLM architecture design, industrialization, training. Comfortable both independently on greenfield projects and as support for existing tech teams.

Industry field of expertise

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Remote only

Primarily works remotely

-
AI Consulting and Delivery for SMEs and Mid-sized Companies
MEDICAL
March 2026 - April 2026 (1 month)
Lyon, France
Context: supporting French companies on their AI projects. Scoping use cases, POCs, LLM architecture design, industrialization, training internal teams. Expertise leveraged: RAG pipelines, LLM agent design, model evaluation and benchmarking, guardrails, hosting choices (public vs sovereign cloud), MLOps.
MLOps LLM Python RAG
-
LLM Benchmark Pipeline
MEDICAL
March 2026 - April 2026 (1 month)
Lyon, France
Context: automated pipeline to evaluate and compare several LLMs on business tasks: test prompt generation, multi-model execution, automatic scoring, reporting. Interest: essential tooling for selecting and monitoring models in production, and adapting each need to the right model.
LLM MLOps Langchain Vector Embeddings Python
-
Multi-Agent System for Material Vigilance Automation
MEDICAL
November 2025 - March 2026 (4 months)
Lyon, France
Context: design and development of a Python multi-agent system to automate the processing of incoming material vigilance emails in a regulated environment (medical sector). The system classifies, extracts structured data, validates compliance, and integrates declarations into the database. Challenges: reliability, traceability of decisions made by agents, patient data security. Achievements: multi-agent architecture, LLM orchestration, automated validation pipeline, database import, audit logs.
LLM Agentic AI Multi-Agent Systems Healthtech Python