Freelancer profile translated to English.

Description

Leveraging my in-depth experience in Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs), I am passionate about innovation and digital transformation within companies. My expertise covers a wide range of advanced technologies and solutions:

🔹 GenAI for Enterprises:

Proficiency in Azure and Microsoft Copilot Studio for integrating optimized and high-performing generative AI solutions.

Deployment of private chatbots for businesses, utilizing advanced RAG (Retrieval-Augmented Generation) techniques on internal documents for intelligent and secure interactions.

🔹 Open Source LLMs:

Expertise in open-source models such as Ollama and Huggingface, with proficiency in LlamaIndex and Langchain libraries for creating custom solutions.

Creation of OpenAI GPTs to meet specific business needs in communication and data analysis.

🔹 LLM Agents and APIs:

Development and deployment of LLM agents via crewAI and integration via REST API for intelligent automation and advanced user interactions.

🔹 Data Engineering and RAG:

Transformation and preparation of multi-type databases and documents for vector databases, facilitating efficient RAG implementation.

Management of data flows to ensure optimal performance and enhanced security.

🔹 Data Security:

Implementation of robust data security solutions, ensuring the confidentiality and integrity of sensitive information in AI environments.

I am dedicated to helping companies navigate the complex ecosystem of generative AI and maximize their potential through innovative and secure solutions. If you are looking for an expert to transform your challenges into opportunities through GenAI and LLMs, do not hesitate to contact me.

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Remote only

Primarily works remotely

Stago
Expert GenAI
BIOTECH
September 2024 - May 2025 (8 months)
Paris, France
Development of a private chatbot based on self-hosted open-source models, enabling interaction with the company's internal documents.

Objectives:
Provide an intuitive user interface for quick access to internal information.
Leverage advanced models for accurate contextual responses.
Technologies used:
RAG (Retrieval-Augmented Generation): for combining document search and text generation.
Ollama: managing LLM models locally.
ChromaDB: vector database for semantic search.
LangChain: orchestrating workflows and integrating models.
Llama 3.2: LLM model for text generation.
Streamlit: interactive user interface.
AWS EC2: cloud infrastructure for hosting.
QA LLM Giskard: tool for model validation and response quality.

Project 2: RAG for Web Scraping
Implementation of a web scraping solution to collect and analyze company-specific information from a list of defined URLs.

Objectives:
Automate the search and collection of relevant information.
Facilitate the integration of data into decision-making processes.
Technologies used:
RAG (Retrieval-Augmented Generation): information retrieval and synthesis.
DuckDuckGo: search engine for collecting external data.
Ollama: integrating open-source models locally.
LangChain Agents: for coordinating automated processes.
Ollama LLM RAG Python N8N
Generali - Europ Assistance
LLM RAG - Teams Bot
September 2023 - July 2024 (10 months)
Project Context:
Development of AI Solutions for Customer Assistance and Internal Management
Athena ChatBot Project
• Design and implementation of an advanced virtual assistant to optimize Europ Assistance's customer service.
• Development of an intelligent conversational interface enabling operators to efficiently handle customer inquiries related to Credit Card Insurance contracts and the Home and Family portal.
• Integration of voice recognition features to improve operational efficiency.
• Application of rigorous evaluation methodologies (PromptFoo, Ragas) during the Prompt Engineering phase to ensure the quality and relevance of responses.

## Deployment of Chatbots on Microsoft Teams

• Adaptation of the Athena chatbot for the Microsoft Teams environment via Copilot Studio.
• Design and implementation of a chatbot dedicated to IT FAQs, facilitating quick access to internal procedures for Europ Assistance employees.
• Optimization of internal workflows through the integration of AI solutions into the Microsoft Teams ecosystem.
Technological Environment:
Python, ReactJS, RAG, unstructured.io, Langchain, Azure Cognitive Search, Azure AI, PromptFoo, Whisper, Microsoft Copilot Studio Teams
Python AI Chatbot RAG Azure Cognitive Services MCP
Galec ( groupe Leclerc)
Lead Dev Python
December 2020 - July 2021 (7 months)
Project Context:
Infra as Code, network automation
Areas of expertise:
Network automation scripts, Infra As Code, automated service deployment on the network using an Azure CI pipeline (opening flows, creating IPs and subnets, creating VMs, creating firewall rules, creating incident tickets). All services are called using Python scripts and their REST APIs.

Technological Environment:
Python 3.9, Azure CI/CD, VMware, Checkpoint, REST API, Swagger