Product · 02 / Computer vision

ProEstimate AI.

Point a phone at a room. Get an AI-rendered remodel and a line-itemed estimate, built for contractors and homeowners.

ProEstimate AI before/after AI render with itemized estimate

The problem

Homeowners want a remodel preview and a ballpark price in seconds. Contractors take days to produce one. That gap is where most remodel leads die. The estimate is the bottleneck — and the bottleneck is mostly visual reasoning a phone can already do.

The product

Open the app. Point at the room. ProEstimate detects the room type, identifies the surfaces and fixtures it can reasonably touch, and produces a render of the proposed remodel plus a line-itemed cost estimate. Every estimate links back to verifiable items priced from regional contractor data.

How it works

  • Capture. The user takes a photo of the room they want to remodel.
  • Generation. The photo, a target style, and a budget tier go to Google's Nano Banana image model via the Gemini API, which returns a remodel render of the same room.
  • Estimation. The detected scope of work feeds a pricing model with regional labor and material adjustments.
  • Delivery. A shareable preview with the before/after, line-item costs, and a follow-up path to a contractor in the area.

The hard parts

Two were genuinely hard: (1) getting Nano Banana to render a remodel of this room — not a stock kitchen — which is mostly prompt engineering and reference framing; and (2) producing estimates that hold up against a real contractor quote, which involves regional pricing tables and tight scope boundaries.

Stack

  • iOS: SwiftUI, native camera capture.
  • Image generation: Google Gemini API — Nano Banana model.
  • Backend: lightweight Python service in front of the model and the pricing tables.
  • Distribution: App Store, with RevenueCat for the subscription tiers.

Status

ProEstimate is shipping on the App Store. Contractor partnerships are being added market by market.

← All products