Fastraise
AI construction assistant.
Core trick
Turns uploaded photos and video into three cost-ranked construction options rendered as a 3D model with Three.js.
Description
AI construction assistant. Upload photos or a video, describe your idea, and explore three ways to build it, evaluating cost, feasibility, by constructing a 3D model, estimating means, materials, risks.
README excerpt
# Fastraise A small construction planning app built with GPT-6 Astra, React, and Three.js. <img width="1486" height="755" alt="image" src="https://github.com/user-attachments/assets/3a805ba2-8d81-45d8-8736-ef521c28c37e" /> ## Run for a demo ```sh npm install cp .env.example .env # only when .env does not already exist # Set OPENAI_API_KEY in .env. npm run dev ``` Open http://localhost:3000. The local key is already configured in this checkout; `.env` is ignored by Git. Hosted deployments use the separate `OPENAI_API_KEY` secret in Sites. 1. The app opens on a clearly labeled illustrative rear-extension project. Switch between the three options and the Existing, Proposed, and Site logistics views. Drag to orbit; scroll to zoom. 2. Click **Use a sample site photo** to load the synthetic example house, or upload your own photos/video. 3. Describe the change, provide a location and an optional measured reference dimension, then click **Explore options** for live Astra analysis. Allow a few minutes. 4. Compare the three approaches, then open **Cost breakdown**, **Build plan**, **People & machines**, and **Assumptions**. **Download plan** saves the analysis and geometry as JSON. 5. **New project** clears the session. **View example** returns to the fixed illustrative example. Example results are never presented as live AI results. The demo photo is AI-generated and is not a real surveyed building. The example's costs are illustrative allowances. Live analysis uses the actu