MY AI BIRD

Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle Submission 💸

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

🐦 Bird Buddy AI — An Open-Weight AI Birdwatching Companion

Bird Buddy AI encourages people to explore nature using open-weight AI. Users can upload bird photographs to receive AI-generated identification suggestions and descriptions of visible features, and take part in a daily outdoor observation challenge.

I built this project to combine AI experimentation with real-world wildlife exploration. It is designed for beginner birdwatchers, nature enthusiasts, students, and educators. In the future, it could grow into a freemium birdwatching companion with advanced identification tools and regional bird guides.

🐦 Try Bird Buddy AI: https://bird-ai-o.streamlit.app/

💻 GitHub Repository: https://github.com/khyatikshitija/bird-ai

I built Bird Buddy AI using Python and Streamlit, with PyTorch and Hugging Face Transformers. Its AI component uses SmolVLM-500M-Instruct, an open-weight vision-language model that analyses uploaded photographs and generates text responses.

The local application runs inference on my computer after the model has downloaded, without sending photos to a separate inference API through the app. I also built a daily nature challenge to encourage users to step outside and observe wildlife.

Building this project helped me learn about model loading, image processing, local inference, and deploying an AI application. I also discovered that general-purpose vision-language models can misidentify bird species, so improving identification accuracy remains an important next step.

Open innovation made it possible for me to experiment with vision-language AI and build a working application without depending on a paid hosted inference API for local image analysis.

Using an open-weight model gave me the opportunity to learn by building, customise the application, and share my code publicly. It also creates opportunities for other developers to contribute improvements, expand species coverage, and make AI-powered nature exploration more accessible.

I'll select the partner prize categories that match the official challenge rules and my project's eligibility.