Bouchra Rguibi

- Developed a full-stack ESG analysis platform using FastAPI, React, and Tailwind, integrating AI and large language models (LLMs) to automate ESG report analysis and comparisons. - Designed a user-friendly interface and efficient backend to support data-driven sustainable finance decisions. - Implemented AI-driven insights to streamline ESG reporting and enhance analysis accuracy.

Built an end-to-end AI system for automated IFRS/ESEF financial report tagging - from raw data to production. - Constructed the training dataset from scratch by extracting and structuring (IFRS tag, accounting label) pairs from raw XBRL files - Built NLP/ML models for automatic prediction of 6000+ IFRS tags - 96% top-3 accuracy - Developed a full React interface - Built a production-ready REST API (FastAPI, API key auth,official IFRS taxonomy integration) - Exposed the entire system via MCP (Model Context Protocol) - enabling AI agents (Claude, Gemini, Cursor) to call the ML models directly in natural language Developed a Microsoft Word add-in for real-time text tagging using LLMs (GPT-4), automating label prediction and improving processing efficiency Worked on a financial Q&A chatbot using Gemini and BigQuery - natural language responses over structured financial data Stack: Python · FastAPI · React · Tailwind · SBERT · scikit-learn · LLMs · Gemini · BigQuery · Google Cloud · pywin32 · FastMCP