BirdLens: Side Hustle

What I Built
I built BirdLens, a simple AI tool that identifies birds from a photo.

The idea came from a simple problem: you see a bird while travelling or walking outside, but you don't know what it is.

With BirdLens, you can upload a bird photo and get the predicted species along with useful information about it.

Demo

How I Built It
I used:

React.js for the frontend
Python + FastAPI for the backend
Hugging Face for the AI model
An open-weight bird classification model for identifying the bird
Why Open-Source AI?
I wanted the AI part of the project to be something I could actually experiment with and control.

Instead of depending completely on a paid vision API, I can use an open model, change it, try other models, and potentially run it locally.

This can also help keep the cost low and gives me more control over how BirdLens works.

For me, that's the main reason open-source AI is interesting: I can build on top of it instead of only using it as a service.

What's Next?
I want to improve the accuracy, add more bird species, improve the UI, and eventually explore offline/local bird identification.

This started as a small project, but I'd like to see if it can become something useful for bird watchers, students, travellers, and nature lovers.

Thanks for checking out BirdLens! 🐦