responsible-ai-evaluation-tools

This project allows data science teams to orchestrate their model governance processes. Ensuring models built for production environments are gated properly with the adequate flags pointing to data and/or model artefacts to review.

A RAG powered compliance screener that assesses AI systems against the EU AI Act using Google Gemini and LangChain. Prototype for educational & research purposes only; not legal advice.

An algorithmic bias audit and fairness evaluation of an emergency department clinical triage model using Azure Machine Learning and macro-balanced mitigation strategies.

Scenario-driven framework for evaluating demographic bias in single- and multi-agent LLM workflows. Tracks bias across sequential stages, separates model non-determinism from systematic effects, and prevents agents inferring evaluation intent. REST API and MCP server.