Health
Occupational Health RAG Assistant
Enterprise RAG
Demo in progressEvidence-grounded assistant for occupational-health documentation, with cited retrieval and answer validation.
- Python
- FastAPI
- RAG
- pgvector
- RAGAS
AI Engineer building practical GenAI and agentic systems
I build practical AI applications combining LLMs, retrieval, agentic workflows and production-ready APIs.
Mobility: Île-de-France · Grand Est
Experience across
Portfolio note
I'm currently connecting live demos and detailed technical case studies.
01Featured projects
Systems built for real problems
Recruitment
AI Talent Matching Platform
Agentic Workflows
Demo in progressAgentic workflow for structuring requirements, ranking candidates and explaining recommendations.
- Gemini
- LangGraph
- FastAPI
- Supabase
- pgvector
Health
Clinical Document Intelligence
Document AI
Demo in progressAI workflow for structured extraction, document comparison and expert-review support.
- LLMs
- Pydantic
- structured extraction
- agentic review
Customer Experience
Real-Time Customer Intelligence
Streaming NLP
Demo in progressStreaming NLP dashboard for detecting sentiment shifts, emerging themes and anomalies.
- Python
- NLP
- embeddings
- clustering
- LLM insights
02Technical capabilities
Core capabilities
Agentic Workflows
- LangGraph
- tool calling
- human-in-the-loop
- automation
RAG & Retrieval
- chunking
- embeddings
- hybrid retrieval
- pgvector
LLM Engineering
- OpenAI
- Gemini
- Mistral
- structured outputs
Backend & APIs
- Python
- FastAPI
- PostgreSQL
- Docker
Evaluation & Reliability
- RAGAS
- LLM-as-a-Judge
- monitoring
- cost & latency
03Engineering approach
From problem to production
- 01→
Understand
Frame the real problem, users and useful outcome.
- 02→
Prototype
Ship a working system end to end first.
- 03→
Evaluate
Measure quality, cost and limitations.
- 04
Productize
Turn what works into reusable components.
Let's build useful AI systems
Interested in practical AI collaborations and production-focused engineering.
Contact me