DIY is a passion for most of the user but the source available in today’s world can’t understand real problem, which leads to confusion, unsafe attempts, or unnecessary expert calls. We wanted AI that can visually understand the situation and guide users confidently.
Our App would help users to build, fix, or assemble things with step-by-step AI visual guidance. When needed, the builders can also connect to real experts and also support the builders in any range of projects.
We built using Google AI Studio with Gemini 3 and Anti-gravity
One challenge was understanding the set-up of the software. Another limitation was adding a broadcast feature, Currently the expert and client needs to be in the same browser due to the absence of a backend, domain, or database.
We successfully built a system that uses real-time visual AI, not just text. We also created a instant broadcast between builder and human experts.
We learned how to use AI live visual supervision and snapshots using Nano Banana. We also learned how Gemini 3 can summarize work, support quick expert help, and turn finished projects into shared achievements and hacks through the Builders Wall.
Next, we plan to integrate Google tools like Calendar, Maps, Weather, and Shopping to make the experience more practical.
Built With
- css
- docker
- google-gemini
- html
- javascript
- react
- tailwind
- typescript
- vite
Log in or sign up for Devpost to join the conversation.