Projects / Supply Chain Risk Analysis
Supply Chain Risk Analysis
Team Lead & Full Stack Developer // Sep 2025 - Apr 2026
Highlights
- Led a 4-member development team from concept to MVP
- 4-stage autonomous AI agent pipeline with LangGraph
- Evidence quality framework with reliability and confidence scoring
- Real-time status tracking via Server-Sent Events
- Deterministic LLM scoring for consistent risk ratings
About This Project
An AI-powered due diligence platform that automates supply chain risk assessments for SMEs, screening suppliers across sanctions lists, debarment databases, Interpol notices, adverse media, ESG metrics, and modern slavery indicators. The platform replaces manual research processes with autonomous agentic workflows that deliver evidence-backed supplier risk ratings.
The system features a 4-stage autonomous agent pipeline (Data Collection, Evidence Analysis, Risk Assessment, Report Generation) using LangGraph for durable, checkpointed execution. It queries multiple public sources simultaneously including international sanctions and watchlists, government debarment registries, Interpol databases, corporate registries, news and adverse media, ESG databases, and company websites.
Built with a Next.js frontend and FastAPI backend, the platform implements an evidence quality framework with reliability ratings, recency scoring, and confidence metrics. Real-time status tracking via Server-Sent Events provides transparent background processing visibility.