SoilSense

Inspiration

Since half of our team calls Ripon home, a town with an abundant amount of farmland, the first problem that came to mind was the lack of water sustainability present in the vast majority of farms which use flooding and mono culture, stripping key elements of the soil and wasting precious water.

What it does

SoilSense is an AI-powered water management system for farms. After gathering weather, soil, and plant health data, the AI is able to make educated guesses on the amount of water an area of the farm will need to thrive while taking into account the costs of buying water.

How we built it

We built this application using HTML, JS, and tailwind CSS. Starting with the frontend, and working our way back to the AI. We also built a simulation powered by our own custom engine, to show how the AI's predictions could affect the health of plants.

Challenges we ran into

One of the biggest challenges we ran into was creating the AI. Due to recent changes with Open AI's API, most of the internet's knowledge is now deprecated, making development difficult and slow. In fact, we are still finishing our code to communicate with our own Assistant.

Accomplishments that we're proud of

This is by far the most complicated and ambitious project that we have undertaken during a hackathon, and the first time we have used AI in one of our application,

What we learned

Around every corner we turned we found struggle GPT problems awaiting us. This taught us the severe importance of resilience as we pushed through every wall that was thrown at us. Additionally, we learned just how tough it is to actually integrate custom GPT’s.

What's next for SoilSense

We would like to better implement our AI and explore more features we could add. Using the AI to also create potential yield and projected water usage amounts and then graph them would be a next step.

Built With

• c++
• css
• html
• javascript
• latnoengine
• node.js