Back to home
Technical documentation

TECHNICAL DOCS

LeadEnrich is an AI-assisted contact enrichment platform. It maps company websites, extracts structured contact data (emails, phones, people), and returns downloadable reports. This page covers architecture and stack details for technical stakeholders.

AI extraction

Uses Firecrawl extract with a typed JSON schema to pull contacts from mapped pages and optional web search context.

Site mapping

Maps each domain (including contact/about/careers style paths) before extraction to improve coverage.

Session quotas

Browser sessions are tracked with a 3-search free limit via X-Session-Id; jobs run asynchronously in the API.

Technology stack

Frontend
Next.js 15 & TypeScript
Backend API
FastAPI (Python)
Extraction
Firecrawl API v2
AI schema
Structured JSON extract
Styling
Tailwind CSS
Proxy
Next.js API route proxy
Concurrency
Async job workers
Deploy
Vercel & Render

How the pipeline works

1. Ingest

POST /api/manual or /api/upload accepts domains, validates the browser session, and creates a background job.

2. Map + extract

Firecrawl /map discovers pages; /extract pulls contacts with schema fields like emails, phones, and representatives.

3. Deliver

Clients poll /api/status/{job_id}. Results export as CSV or Markdown via download endpoints.

H
Lead AI Engineer

Huzaifa Abdurahman

Lead AI Engineer · Apexa AI Labs

As Lead AI Engineer at Apexa AI Labs, Huzaifa designs production AI systems for enrichment, automation, and intelligent data workflows — helping teams turn web data into actionable sales and growth outcomes.

Apexa AI Labs

Building practical AI software

Visit apexaailabs.com