Case study

FixPoint

Property maintenance software that lets AI triage tenant repair requests and draft contractor-ready work orders.

Role
Product + marketing
Year
2026
Engagement
Independent
next.jsvision-aipropertyproptechmaintenance

Stop being the middleman for every repair.

FixPoint takes the landlord out of the switchboard. Tenants report a problem through a no-app chat link; the AI triages urgency, reads submitted photos to pull appliance model numbers and assess damage, and assembles a work order a contractor can act on — all in one shared thread.

We built it on Next.js with a vision-and-language pipeline doing the triage. Landlords approve and dispatch instead of relaying messages, and tenant, owner, and contractor all see the same thread.

Highlights

  • No-app reporting — Tenants report issues through a simple chat link — nothing to download or log into.
  • AI photo analysis — Vision reads tenant photos to extract appliance model numbers and gauge the damage.
  • Urgency triage — Every request is auto-rated green, amber, or red so the real emergencies surface first.
  • One shared thread — Tenant, owner, and contractor stay in a single thread — the landlord never plays switchboard.

Role

Product and marketing build — Next.js, vision + LLM triage pipeline, Vercel.

Website: fixpoint.believelabs.com

More work

Show all work
Available for consulting

Have something to build?

I take on a small number of consulting engagements each year. Tell me what you're working on — I reply within a day or two.