Freelancer profile translated to English.

Description

I support companies in the design and production of Generative AI and agentic AI systems, capable of automating complex processes and generating measurable business impact.

With 6+ years of experience in AI (Machine Learning, NLP, LLM, RAG), I design solutions that go beyond simple chatbots, towards truly autonomous intelligent agents.

My key expertise:

- Design of autonomous AI agents (multi-step, tool usage, reasoning)

- Integration of LLMs in production (RAG, fine-tuning, orchestration)

- Agentic systems connected to IS (APIs, databases, business tools)

- Industrialization via MLOps / LLMOps

- Monitoring: hallucinations, drift, performance, costs

- Consideration of security, compliance, and governance issues

Use cases I develop:

Advanced automated customer support agents

Fraud detection and document analysis agents

Augmented business assistants (finance, insurance, ops)

Automation of complex workflows (multi-source, multi-action)

RAG systems for internal data valorization

My added value:

I transform POCs into reliable, scalable, and truly production-ready AI systems.

End-to-end intervention:

Scoping → Architecture → Development → Deployment → Monitoring → Optimization

📩 Available for high value-added missions

→ AI agent build, LLM integration, production scaling

Industry field of expertise

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km)

BPCE
IA Architect & MLOps
BANKING AND INSURANCE
April 2024 - Today (2 years and 2 months)
- Generative AI solutions architecture.
- Production deployment
- MLOPS / LLMOPS
- Multi-agent systems
- FRAUD DETECTION
LLMs Machine learning RAG Google cloud FastAPI
AXA
Senior ML Engineer
BANKING AND INSURANCE
November 2021 - Today (4 years and 7 months)
Paris, France
- Design and production deployment of large-scale Machine Learning systems in the insurance sector, aiming to optimize operational processes and improve customer experience.

- Development of an LLM text analytics solution for customer feedback analysis, contributing to a +15% NPS improvement.

- Architecture of an advanced fraud detection system combining supervised and unsupervised models to improve fraudulent claim detection and reduce financial losses.

- Integration of Machine Learning models into business decision-making systems (pricing, claims management) to automate decisions and improve operational efficiency.

- Industrialization of models via an MLOps platform based on Databricks and MLflow: experiment management, versioning, automated deployment, and performance monitoring.

- Collaboration with Fraud and IT teams for the industrialization and operational adoption of ML solutions.

Technologies: Python, SQL, PySpark, Azure, AzureDevOps, AzureML, Databricks, MLflow, Docker, LLM, CI/CD, MLOps
Python PySpark Machine learning DevOps Azure Databricks
BNP Paribas Personal Finance
Data scientist
BANKING AND INSURANCE
April 2020 - September 2021 (1 year and 6 months)
Paris, France
- Challenging and improving existing credit scoring models by implementing Artificial Neural Network (ANN) approaches to evaluate and enhance predictive default risk performance.

- Comparative analysis and optimization of model performance to improve default risk prediction and strengthen the robustness of credit decision systems.

- Participation in the integration of analytical models into credit decision systems, in collaboration with data and business teams.

Technologies: Python, TensorFlow, Scikit-learn, Git
Machine learning TensorFlow Credit risk Credit scoring Deep Learning