May to Jun 2026
Recon
An AI agent for data center site selection.
A full-stack agentic app that chains tools over zoning ordinances and public infrastructure data to evaluate where a data center could go.
Illustration, not real data. Fictional terrain.
The five public sources Recon reads, as map layers. Toggle them.
No screenshot yet. The live deployment is being fixed; a real capture goes here once it is back up.
- recon › evaluate candidate site B
- the agent picks the next tool
- → zoning ordinancesRAG search
- → FEMAflood zones
- → HIFLDinfrastructure
- → USGSgeology
- → EIAenergy
- → Mapboxbasemap
- → SQLite memoryrank against prior sites
- done. outputs omitted here
Trace shape, illustrated. The agent chooses the order; outputs are left out on purpose.
The idea
Picking a data center site means reading zoning ordinances, checking flood risk, power, geology, and infrastructure, then remembering how previous sites scored. Recon turns that into an agent loop with a live tool-call trace.
When an agent chains five APIs, you want to see which tool ran and why, not just the final answer.
The build
- 1An LLM loop that selects and chains tools.
- 2RAG over zoning ordinances.
- 3SQLite memory that ranks new sites against prior evaluations.
- 4Five public-data APIs: FEMA, HIFLD, USGS, EIA, and Mapbox.
- 5A live trace UI so you can watch every tool call.
What happened
A working agentic pipeline for site evaluation. The live deployment is being fixed, and a corrected link is coming soon.