Old McDonald

Inspiration

The inspiration for Pocket Farmer came from the growing trend of urban farming and the observation that its potential is hampered by a high initial learning curve. We saw a disconnect between the rising interest in urban agriculture and the overwhelming amount of information required to be successful. We wanted to empower aspiring urban farmers, regardless of their experience, to cultivate thriving green spaces in their communities.

What it does

Pocket Farmer is a mobile application that serves as a comprehensive urban farming assistant. It features three key functionalities:

Plant Oracle: This location-based feature utilizes user location, real-time weather data, and seasonal trends to suggest plants with the highest success potential for their specific urban environment. Machine learning analyzes historical agricultural data and environmental factors to provide hyper-local recommendations, taking into account unique microclimates and limitations of space.

Plant Identifier: This feature utilizes image recognition technology. Users can upload pictures of unknown plants for identification. This allows them to research proper care requirements and make informed decisions about integrating these plants into their urban farms.

Plant Diagnostics: This feature facilitates the identification of plant illness. Users upload photos of sick plants, and the app, through machine learning and image analysis, suggests potential diagnoses and provides actionable remedies to heal their crops.

How we built it

NextJS Rust Vertex AI Tailwind CSS GCP (Cloud Run Container)

Challenges we ran into

Government weather API was hard to work. There was also not enough time to implement as many GCP features as we would have wanted to. Lastly, there were some technical challenges when trying to deploy to GKE so we decided to go with Cloud Run instead.

Accomplishments that we're proud of

Developing a user-friendly mobile application that empowers aspiring urban farmers.
Creating a location-based recommendation system using machine learning to provide hyper-local plant suggestions.
Integrating image recognition technology for plant identification and disease diagnosis.

What we learned

The importance of making complex agricultural knowledge accessible and user-friendly.
The power of machine learning in providing personalized and data-driven recommendations.
The value of image recognition technology in assisting with plant identification and health monitoring.

What's next for Old McDonald

Expanding the plant database to include a wider variety of crops and ornamentals.
Developing a community forum within the app to foster communication and knowledge sharing among urban farmers.
Integrating with smart gardening tools to provide automated watering and environmental monitoring.