Freelancer profile translated to English.

Description

I design and deploy LLM systems in production, from research paper to tested and delivered service in a few weeks. If you have an AI use case (agents, RAG, data anonymization, prompt optimization) to bring to production, that's my domain.

What I bring you concretely:

→ a service that runs, protected by automated tests and quality gates

→ a solid Python backend foundation (FastAPI, Docker, GCP): I deliver the AI layer and everything around it

→ a focus on cost and reliability: budget tracking in euros, telemetry, model evaluation and benchmarking

What sets me apart: I deliver quickly and cleanly. At my last client (an agent platform for live music), I built a PII anonymization service from scratch to a production quality gate in 13 days.

Types of projects / deliverables: MCP and REST services, PII anonymization pipelines, prompt optimization and LLM evaluation, data pipelines (Databricks, Azure), custom LLM integrations.

Bilingual FR/EN. Paris and remote.

Industry field of expertise

Languages

French
Native or bilingual
English
Native or bilingual

Workplace preferences

Can work on-site

Paris 20 Ménilmontant (up to 50km)

2AM
AI/ML R&D Engineer
TECH
January 2026 - July 2026 (6 months)
Paris, France
→ Developed a stateless PII detection and anonymization service (38 types covering 4 languages, regex + LLM), serving as a privacy barrier for an agent platform. From scratch to a production quality gate in 13 days.
→ Built a prompt optimization pipeline from current research and tooling, pushed further and packaged into a production MCP and REST service, with euro budget tracking and live telemetry. From paper to deployment in about three weeks.
→ Approximately 8,500 lines of production Python across both services, both initialized from my own public template and protected by automated quality gates.
MCP Python Artificial Intelligence Back-End Development Research and Development (R&D)
Eviden (DataSentics)
ML / Data Engineer
TECH
June 2023 - September 2024 (1 year and 3 months)
Paris, France
→ Design and optimization of scalable data pipelines on Databricks and Azure
→ LLM Project: conversion of natural language to SQL queries, prompt engineering, and fine-tuning
→ Close collaboration with clients for complex dataset analysis
→ Databricks Data Engineer Associate Certification
Artificial Intelligence Python Research and Development (R&D) Databricks Microsoft Azure
STMicroelectronics
Software Engineering Intern
TECH
June 2019 - January 2022 (2 years and 7 months)
Silicon Valley, CA, USA
Improved, debugged, and created automation software for Imaging Department VCSEL Team testing using LabVIEW and Python. Automated lab equipment integration. Verified efficiency and stability of VCSELs (Vertical Cavity Surface Emitting laser) used for 3D sensing. Sped up testing times for VCSEL tests ultimately allowing the VCSEL team to be more productive.
Python Research and Development (R&D) Computer Vision Labview Automation