Description

ML engineer with 6+ years of experience designing and deploying production GenAI systems: RAG pipelines, LLM fine-tuning, ML APIs, and evaluation infrastructure.

I cover the full stack — from data pipelines and model training to deployed APIs and production monitoring.

Core expertise:

— RAG systems (LangChain, LangGraph, vector search, Elasticsearch)

— Domain-specific LLM fine-tuning

— LLM evaluation and tracing with Langfuse

— GenAI APIs on AWS (EC2, S3, SageMaker)

— Multi-agent workflows (LangGraph, MCP, n8n)

Recent projects: full GenAI feedback analysis system at Feedier (Lille), production recommendation engine at Nextory (Paris).

Available for freelance missions in Île-de-France and fully remote.

Immediately available.

Domaines d’expertise

Langues

Anglais
Bilingue ou natif
Français
Capacité professionnelle limitée

Préférences en matière de lieu de travail

Accepte de travailler sur site

Paris (jusqu’à 50 km)

Feedier
ML/AI engineer
octobre 2024 - janvier 2026 (1 an et 3 mois)
Lille, France
Designed and deployed LLM-powered systems for customer feedback analysis, including information extraction, summarization, and insight generation.
Built ML models (clustering, classification, sentiment analysis) to enhance LLM outputs. Implemented RAG pipelines with Elasticsearch and vector search to deliver contextual, high-precision responses.
Fine-tuned domain-specific LLMs, improving relevance and accuracy for customer-experience use cases.
Architected stateful agents using LangGraph and MCP servers, ensuring production reliability via LangSmith tracing.
Ran systematic LLM benchmarking and evaluations focused on accuracy, latency, and reliability across model families.
Architected and deployed AI-driven automations using n8n, integrating LLMs and external APIs to streamline operational workflows.

Tech Stack: Python, PyTorch, TensorFlow, scikit-learn, langChain, Langfuse, LiteLLM, Elasticsearch, MySQL, RAG, MCP Server, LLM fine-tuning, Prompt Engineering, EC2, S3
Python LLM artificial intelligence RAG AI Automation
Nextory,
Data Scientist (Recommendation Engine)
DIVERTISSEMENTS & LOISIRS
septembre 2022 - septembre 2024 (2 ans)
Paris, France
Built and deployed ML models for user behavior prediction and personalized content ranking, contributing directly to engagement and recommendation quality.
Developed and maintained Flask-based APIs powering the recommendation engine, and managed data storage across BigQuery, PostgreSQL, and MongoDB.
Engineered a hybrid retrieval system combining vector search (embeddings) and string-matching to deliver highly relevant content recommendations.
Applied strong Python and ML expertise (scikit-learn, PyTorch/TensorFlow) for model development, evaluation, and iterative improvement.
Python Machine learning Data science Gestion de base de données (SQL, NoSQL, etc.) artificial intelligence
NAISTER
Data Scientist, (NLP/Machine Learning)
mars 2019 - août 2022 (3 ans et 5 mois)
Paris, France
Leveraged AI/ML to develop an impactful Content Strategy with cross-functional teams. Direct experience with state-of-the-art LLM architectures, including GPT-3 and Transformer-based models (BERT, RoBERTa).
Deployed and trained deep neural networks using AWS SageMaker for critical scoring applications.
Built a Video Text Retrieval system, aligning with Information Retrieval requirements.
Developed high-impact Ranking and Recommendation Systems for e-commerce.
Experience with AWS cloud infrastructure, including SageMaker, Lambda, DynamoDB, S3, and EC2 for scaling ML solutions.
AWS SageMaker Machine learning