Description

AI Engineer specialized in LLM applications, Retrieval-Augmented Generation (RAG), and intelligent automation.

I design and deploy AI systems that transform data into actionable insights and scalable products. My expertise includes building LLM-powered applications, AI chatbots, and data pipelines for real-world business use cases.

Currently working as an AI Engineer intern at Institut Pasteur, I develop machine learning solutions applied to biomedical and speech datasets, combining advanced modeling with production-ready engineering.

Core expertise:

• Large Language Models (LLM) integration

• Retrieval-Augmented Generation (RAG)

• Vector Databases (FAISS, Pinecone, Chroma)

• AI Chatbots & Conversational Agents

• Python / Machine Learning / NLP

• Data pipelines & automation

• API integration and backend systems

Tech stack:

Python • PyTorch • TensorFlow • LangChain • OpenAI API • Hugging Face • Docker • Git • SQL

What I can help you build:

• AI chatbots and assistants

• Knowledge-base search systems (RAG)

• AI automation workflows

• Custom ML/NLP solutions

• Data analysis and predictive models

If you're looking to integrate AI into your product or automate complex processes, I can help you design robust and scalable solutions.

Langues

Français
Bilingue ou natif
Anglais
Bilingue ou natif

Préférences en matière de lieu de travail

Accepte de travailler sur site

Paris (jusqu’à 50 km)

Institut Pasteur
AI Engineer – Medical AI Systems
CENTRES DE RECHERCHE
mars 2026 - Aujourd'hui (5 mois)
Paris, France
- ENTIA medical chatbot: developed a conversational assistant for ENT physicians (RAG/LightRAG) with multi-agent orchestration (router + specialized agents), a knowledge graph, and BGE-reranker-v2-m3
- ORL-Bench — selecting the best model & RAG for ENTIA: full-stack platform (React, NestJS, FastAPI) to train, test, and compare LLMs on 600 clinical cases; goal: choose the optimal configuration to integrate into the ENTIA product
- Evaluation & POC: RAG pipeline (FAISS, embeddings, jina-reranker); comparison of 3 strategies (direct / RAG / RAG+rerank); automated KPIs (scientific_score), export to JSON/PDF/JSONL; validation on 30 cases (100% pass rate)
- Multi-LLM integration: connected multiple models (Mistral, Gemini, DeepSeek, Qwen, OpenAI) via LiteLLM; prompt engineering and performance comparison to guide deployment decisions
- MLOps & industrialization: Docker/Kubernetes, GitLab CI/CD (Kaniko/Harbor), Redis/RQ async jobs, JWT auth, circuit breaker; technical documentation and feedback sessions with domain experts
Vector DB knowledge base search Langchain Retrieval-Augmented Generation (RAG) Chatbot IA
OCP & YouCode
Full-Stack Developer / Data Analyst
GRANDE DISTRIBUTION
mai 2024 - septembre 2024 (4 mois)
Morocco
• • End-to-end application design using UML, wireframes and system architecture
• • Web development using React and Node.js
• • Implementation of REST APIs and MySQL databases
• • Automation of data pipelines connecting APIs, databases and Power BI
• • Development of decision dashboards with business KPIs
• • Collaboration using Git/GitHub in Agile environment
React.js Node.js Python JavaScript MySQL
OCP Group
AI Data Analysis Pipeline - Predictive Maintenance
INDUSTRIE MATIÈRES PREMIÈRES
juin 2023 - septembre 2023 (3 mois)
Morocco
• • Data preprocessing and analysis using Pandas and NumPy for maintenance my purposes.
• • Feature engineering and predictive modeling in Python
• • Implementation of Random Forest and Logistic Regression
• • Model evaluation using statistical performance metrics
• • Data visualization using Matplotlib
Microsoft Power BI Random Forest Machine learning Deep Learning Processus ETL (Extract, Transform, Load)