Freelancer profile translated to English.
Back to original languageAbout Maxence
AI Developer specializing in LLM integration in production.
I design and develop artificial intelligence-based applications for companies that want to go beyond POC. My expertise covers LLM API integration (OpenAI, Claude, open-source models), building RAG systems, and fine-tuning NLP models.
Key Skills:
LLM Integration:Connecting APIs (GPT, Claude...) to your business applications. Conversational agents, automated workflows, intelligent user interfaces.
RAG Systems:Building assistants that leverage your internal documents. At Enedis, I deployed a RAG system processing 500+ technical documents with an extraction pipeline and semantic search.
NLP & Fine-tuning:Training specialized models for your use cases. At Enedis, I fine-tuned BERT for named entity recognition based on a business ontology. Stack: PyTorch, Hugging Face, LoRA/PEFT.
Audit & Optimization:Analyzing existing ML pipelines to identify performance issues, implementation bugs, or improve results. Methodical approach and concrete recommendations.
Technical Stack:Python, PyTorch, Hugging Face, FastAPI, LangChain
Profile:Software Engineer (Epitech) + AI Master's Degree. I prioritize maintainable production code over academic experimentation.
Format:Short missions (1-4 weeks), remote or Marseille.
French
Native or bilingual
English
Fluent
Can work on-site
Toulon (up to 50km), Marseille (up to 10km)
Experience
- Enedis – Laboratoire d'Intelligence Artificielle,AI InternENERGY AND UTILITIESOctober 2024 - August 2025 (10 months)Marseille, France
- Fine-tuning LLMs with GRPO (Group Relative Policy Optimization) under structural constraints
- Integration of a Prolog-based reward system to guide grammar-compliant generation
- Comparative evaluation of graph generation techniques
- Enedis – Laboratoire d'Intelligence Artificielle,AI InternENERGY AND UTILITIESApril 2024 - July 2024 (3 months)Marseille, France
- Design of a business ontology and a PEG grammar to constrain the generation of interpretable graphs
- Development of an NER pipeline for entity extraction in technical documents
- Implementation of a RAG (Retrieval-Augmented Generation) system for HR documentation
- Overall objective: automatic generation of a knowledge graph from unstructured text
- Intermed ExportationAI / Data Science InternIMPORT AND EXPORTJuly 2023 - February 2024 (7 months)Hyères, France
- Training a BERT model to identify similarities between drug names
- Creation of a dashboard to visualize B2B commercial anomalies
- Development of document extraction tools with the OpenAI API
Recommendations
FU
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
- Master in Artificial IntelligenceCenter for Research and Teaching in Computer Science - University of Avignon2025
- Expert in Information TechnologyEpitech2024Ecole d'informatique en 5ans qui à pour objectif de former ses étudiants aux métiers du numérique.