Policy Intelligence (PI) Lab

Deep Learning for democratic and equitable governance

PILab is UEI’s applied research and public-interest technology unit, dedicated to building tools that make complex policy landscapes clearer, more accessible, and more democratic. We treat policy analysis not as a static academic exercise, but as a dynamic, computational challenge that demands new methods and new imagination.

At PILab, we design and deploy systems that read, interpret, and visualize policy in real time—from federal appropriations bills to local ordinances shaping the lives of migrants, tenants, and city residents. Our work bridges political science, urban governance, civic technology, and machine intelligence, giving researchers and policymakers the ability to see across documents, detect hidden patterns, and understand the structural consequences of legislative change.

Our flagship project—the evolution from a simple Retrieval-Augmented Generation (RAG) tool into a full Intelligent Retrieval (IR) engine—captures what PILab stands for: rigorous engineering married to real civic purpose. By combining vector databases, LLM reasoning, and clean, intuitive interfaces, PILab builds policy tools that work for people, not just experts. PILab is driven by a mission to democratize policy knowledge, empower communities, and push the frontier of student-built public-interest tech. We prototype fast, test ambitiously, and always aim to produce insights that matter for cities, institutions, and the communities they serve.

Current Projects

AI-Enhanced Policy Analysis Through RAG/Intelligent Retrieval - LLM-embeds to vector database, with front-end querying for complex, incisive analysis on the most recent appropriations bill (H.R.5371). Enabled through VS Code, Gemini 2.5-flash, and Pinecone.

Building a RAG-Intelligent Retrieval Policy Analysis Engine for Congressional Legislation

Over the past term, the Policy Insight Lab (PILab) has pushed beyond conventional research workflows to build a new class of policy analysis tools. What began as a simple Retrieval-Augmented Generation (RAG) prototype—intended to extract facts from complex federal documents—has evolved into a dynamic Intelligent Retrieval (IR) engine

The Urban Equity InstituteIshaan Barrett

Classification and Clustering of City Governance Reforms - PCA clustering and sentiment analysis of NYC Policy and Procedural Recommendations (PPRs) to identify patterns in reform activity and investigations at the Mayoral level. Enabled through Gemini and Gemini Colabs.

PCA Clustering and Topic Modeling of New York City PPRs Demonstrate Typological Differences in Content and Success Rate of DOI Interventions in City Governance

This research investigates policy and procedural recommendations (PPRs) in New York City governance using PCA cluster analysis and natural language processing (NLP).

The Urban Equity InstituteThe Urban Equity Institute