RevvIQ
Legal intelligence platform that scrapes, wrangles, and matches claim and court data across NYOSC, Westlaw, and eCourts — with an AI agent that turns natural-language questions into SQL.
Business Problem
Legal and claims teams need actionable intelligence from fragmented public sources — NY Workers' Compensation claims (NYOSC), Westlaw research, and New York eCourts dockets — but each portal is siloed, hard to monitor at scale, and impossible to join without heavy manual work. Clients also need keyword-driven discovery and a way to query the resulting warehouse in plain English.
Solution
Built RevvIQ as an end-to-end data platform: scrapers for NYOSC claim search, Thomson Reuters Westlaw, and NY eCourts; Python pipelines for preprocessing, wrangling, extraction, and analysis; address-based record matching and client-keyword scrape orchestration; a React frontend with Resend for transactional email; and an AI conversational agent that generates SQL from user questions and returns live results. Deployed on Render.
Tech Stack
- Python
- SQL
- React
- Web Scraping
- Data Wrangling
- LLM
- Text-to-SQL
- Resend
- Render
Impact
Unifies multiportal legal and claims data into a queryable, matched dataset — so teams can monitor keywords, link records by address, and ask natural-language questions against the warehouse instead of stitching portals by hand.