InfillOS / Our product / 2026 to present
Finding infill lots that pencil, before anyone else does
A feasibility engine covering 132,562 Durham, NC parcels, with a founding developer who is also its investor.
The problem
Small residential developers find infill opportunities one lot at a time. Checking a single parcel's zoning, fitting a building to it and running the numbers can take days, so most good lots are never evaluated at all.
What we did
We ingested every parcel in Durham, North Carolina, 132,562 of them, and resolved the zoning constraints that apply to each one.
The hardest problem was fitting buildings onto irregular lots. We researched and implemented a hybrid genetic-algorithm and simulated-annealing solver that packs units onto odd shapes, then runs a pro forma and scores every result.
Instead of building one more dashboard, we exposed search, evaluation and mapping as MCP tools, so a developer can ask for viable lots from whatever AI assistant they already use. A web dashboard handles maps and review.
The result
InfillOS is running in Durham as its first market, with invite-only access.
BRM Collective, a land developer, is the founding customer on a profit-participation model and is also InfillOS's first investor.
Need this?
- Agentic AI & RAG builds
We design and ship AI assistants, agents and MCP servers that run in production, not just in a demo.
Related work
- AEOifyOur product
Building an AEO platform that brands pay for
A production AEO SaaS now used across Conair Corporation brands and by Gordini USA.
AEO & e-commerce growthAgentic AI & RAG builds
- Portland Old PortClient work
A production AI assistant for Portland's Old Port
A retrieval-augmented assistant that answers visitor questions from the district's own content.
Agentic AI & RAG builds
- Maine residents and boatersOpen source
PublicMCP: public services as a conversation
An open-source MCP server for Maine burn permits, fire danger, tides and marine forecasts.
Agentic AI & RAG builds