Inspiration
The current LLM is not specific enough for health-related problems. So we are introducing a health-oriented AI agent
What it does
By using API to use Gemini to answer the questions, we used the templates to regulate the answer.
How we built it
Based on the previous framework, we make the code more specific and health oriented.
Challenges we ran into
Due to the specific proxy and internet environment, the usage of API was kimited at the beginning. Anyway we fixed that in the end.
Accomplishments that we're proud of
The demo question for the glucose level problem works well in this project.
What we learned
API can be a good tool to build specific models. And it is wise to use some templates for the output.
What's next for Health Agent Based on LLMs
Optimization for the output. We may have more codes to doule-check the output to make sure it is accurate and responsible.
Log in or sign up for Devpost to join the conversation.