Averom — AI Supplier Intelligence Platform

Built for Amazon FBA sellers and B2B operators who need to verify suppliers before placing purchase orders.

Problem

Counterfeit suppliers and sourcing fraud cost importers millions every year. Manual vetting is slow, expensive, and unreliable.

Solution

Averom uses Claude AI to autonomously research suppliers, score risk across 7 weighted categories, and generate structured intelligence reports in seconds.

Features

• AI Supplier Verification Agent
• Deal Profitability Analyzer
• Sourcing Engine with AI-powered matching
• Multi-turn AI Copilot with live data context
• Communication Hub for context-aware outreach
• Deal Pipeline with RAG-grounded responses

Tech Stack

• Frontend: Next.js, TypeScript
• Backend: Vercel Serverless Functions
• Database: Supabase (PostgreSQL + pgvector)
• AI: Claude API (Anthropic) with web search tool
• Payments: Stripe
• Deployment: Vercel

Architecture

User Input → Next.js Frontend → Vercel API Route → Claude API (with Supabase context injected) → Structured Output → Supabase Storage → Dashboard

Live Demo

• Platform: https://averom-v2.vercel.app
• Supplier Verification: https://averom.co

Why This Is Not Just an AI Wrapper

Averom is not a thin wrapper around a language model. The system includes:

• Deterministic risk scoring across 7 weighted supplier categories before any AI call
• Structured prompt engineering that injects live Supabase context (past reports, deal history) into each Claude request
• pgvector semantic search for RAG-grounded supplier matching, not keyword lookup
• Agentic web research using Claude's web search tool to pull real-time supplier signals
• Persistent deal state stored and retrieved across sessions, not re-generated each time

The AI layer handles reasoning and synthesis. The deterministic layer handles scoring, routing, and data retrieval. Both are required for the system to function.

Execution Speed (built in 1 week, 30+ reports, 15 waitlist)

• Built from zero to deployed platform in under 1 week
• 30+ supplier intelligence reports generated in production
• 15 businesses on the waitlist before public launch
• 8 agentic modules shipped: verification, deal analysis, sourcing, copilot, outreach, pipeline, comparison, and risk scoring