regulatory-ai

Multi-agent AI framework for EU policy, law, and institutional procedure. Built for Brussels, open to the world. EU Prompts : https://montoyer.github.io/eu-agents/

AI compliance copilot using Retrieval-Augmented Generation to guide entrepreneurs through UAE free-zone company setup workflows.

Citation-first regulatory AI system for Delhi Electricity Law (DERC), built with a reproducible document pipeline, strict refusal rules, and governance-focused LLM design.

Agentic AI prototype for commercial decision intelligence in life sciences — translating market signals into competitive, strategic, and regulatory recommendations in real time.

An AI system that evaluates consumer complaints and corresponding company responses to identify potentially inadequate, dismissive, or misleading complaint resolutions.

CABS 2026 intern project - OpenRegEval (Matías Pinto)

Applied AI project using Mistral, TCN, QLoRA baselines, and PyTorch for financial and regulatory text classification

Toy credit-decisioning agent with tamper-evident audit log and declarative policy guardrails. Explores how to make LLM agents auditable for regulated industries.

Administrative decision-support system for prior authorization readiness. Deterministic, rules-first evaluation of documentation completeness with refusal semantics, policy provenance, and full auditability. Built for safety, transparency, and regulated healthcare workflows.