Inspiration
The inspiration for the project was honestly us having a conversation before the hackathon started about how two of our team members had mismanaged their time and done horrible on a test.
What it does
Better Brain is your own personal assistant at the tip of your fingers. It uses an ever-growing well of knowledge to customize your schedule around your needs and preferences, allowing you to focus on what really matters, LEARNING.
How we built it
We had to really research prompt engineering and vector databases, as that is what allows communication with Better Brain to be so seamless. We used Python and Llangchain to build an environment where Ollama's 3b llama 3.2 model could pull from different sources provided by the developers and could understand a users needs and through clever prompt engineering we could create a simple web API that would allow the user to interact with the product.
Challenges we ran into
Figuring out a way to make our output structure consistent was quite a challenge, as the limited complexity of the llm available to us meant that results could vary drastically. Another challenge was coming up with a nice, simple, but comforting UI.
Accomplishments that we're proud of
We powered through all the painstaking challenges and came out with a good product with tons of headroom for further improvement.
What we learned
We learned a lot about managing expectations and maintaining a positive attitude throughout all the struggles that come with developing a product like this.
What's next for Better Brain
Better Brain may very well se further development and may become a real product with time.
Log in or sign up for Devpost to join the conversation.