Built an AI-powered fleet management assistant using Google ADK and a custom Model Context Protocol (MCP) server, enabling natural language queries over live EV charging infrastructure data. Designed an automated benchmarking pipeline to evaluate AI assistant accuracy and response quality.

Part-time AI Engineer

NartaQ · Paris, France

October 2025 – January 2026

Developed a startup–investor matching system using Azure OpenAI embeddings to generate semantic representations of structured and unstructured data, improving matching accuracy by 30%.

3 more items available to recruiters