AI Engineer · M.Sc. Data Science & AI @ EPITA (SCIA)
I build and ship production GenAI systems — agentic workflows, multi-agent, knowledge graphs, retrieval. Most of my work lives in the layer above the model: getting LLMs to reason, use tools, retrieve the right context, and do useful work reliably. I like owning a project end-to-end, from architecture to deployed product.
- 1st place — HrFlow.ai GenAI & HR Hackathon (2026): an autonomous sourcing agent that returns scored candidates with SWOT analysis and outreach drafts in under 2 minutes.
- Finalist — Hi!Paris × HEC × Polytechnique, AI & Education track.
- Generative AI Engineer Intern @ Synechron: designed a RAG + agentic code-generation PoC, adopted internally as a FinLab tool and now moving to production.
- haunt — AI phantom users that test your app the way real users actually use it.
- arch-explainer-code — Drop any codebase, get an instant architecture map plus a RAG-powered chat that cites your actual source files.
AI / LLM — Python · LangChain · LangGraph · RAG · GraphRAG · multi-agent · PyTorch Data — PostgreSQL · Neo4j / Cypher · Weaviate · Pandas · NumPy Build — TypeScript · Next.js · FastAPI · Docker