DevelopIQ is an AI-powered tool for real estate developers to instantly surface zoning and development constraints — starting with automated site investigation reports.

We help developers, brokers, and land teams stop wasting hours on PDF scavenger hunts and city planner calls.

What’s working today — and where it delivers value.

Given an address or parcel ID, we fetch and aggregate:

• Parcel metadata (location, size, owner, use class)
• Market-level data (job trends, housing supply, demographics)

🔍 Data sourced from public APIs and government databases.

This is our LangGraph-powered agentic system, focused on zoning feasibility.

• 🗂️ Extractor Agent

  • Finds and retrieves the appropriate municipal codebook (HTML or PDF)
  • Chunks and indexes it for semantic search

• 🔍 Querier Agent

  • Asks predefined development questions:
    • What are the permitted uses for this site?
    • What are the setbacks and height limits?
    • What is the rezoning process?
    • Are there historical overlays or design commissions?
  • Returns answers (with source citations) based on zoning code

🚨 This is not a chatbot. It’s a deterministic pipeline focused on structured zoning answers.

/frontend          ← Next.js (React, Tailwind) app for UI
/codebook_agent    ← LangGraph backend (agents + codebook parsing logic)
/util_python_lib   ← Utility scripts (data ingestion, API pipelines, needs cleanup)

• Hosted on LangGraph Platform
• Data currently flows one-way from static sources
• No user-upload support yet (coming)

cd frontend
npm install
cp .env.example .env.local
npm run dev

Frontend runs on http://localhost:3000

Use Langsmith or LangGraph local dev mode.

langgraph dev

• ❌ No AI-generated proformas or valuations (yet)
• ❌ No user-uploaded documents (yet)
• ❌ No live code conflict detection (yet)

This is MVP 0.1 — validating structured site investigation automation.

• Integrate parcel lookup with zoning agent output
• Add surface-level red flag detection (e.g. “requires rezoning”, “2-year entitlement”)
• Merge historical overlays and flood/topo layers
• Create downloadable SIR (PDF or markdown)

• Kush Bhuwalka (rhit-bhuwalk)
• Logan McLaughlin
• Evan Brooks

MIT License