Coinspector — Blockchain Forensics
Cryptocurrency forensics platform built at Lab Systems for tracing illicit on-chain activity. Distributed RPC ingestion, ML-based clustering heuristics, and a sleek investigator UI shipped to forensic clients.
Coinspector is a specialized investigator tool for cryptocurrency fraud — tracing transaction graphs, clustering wallets, and surfacing suspicious patterns on Bitcoin and EVM-compatible chains. I built the platform end-to-end at Lab Systems: a NextJS + TailwindCSS interface, a Python ML analyzer consuming a Kafka transaction stream, and MongoDB persistence served through a Vercel edge cache.
Throughout development, I implemented testing harnesses to validate clustering accuracy and data provenance, ensuring investigators received trustworthy, reproducible insights. Coinspector pairs cutting-edge analytics with a clean operator experience and remains in active client use as part of the firm's forensics suite.
Project Details
Technical Insight
This project utilizes a high-performance distributed architecture and specialized AI integration.