Inspiration
Being able to easily look for different plans that have meet different criteria. I have someone special in my family and sometimes is hard to find the right plan give different conditions
What it does
It helps users find the appropriate plan for them that meets their needs.
How we built it
We use the data set provided and we extracted only tables that were tailor to special need type plans, then we created custom functions that allow simple searches through the dataset. This functions would help the model know what to look for a given question. Then we feed the data and the functions to llama 3.3 to help with the langchain thinking behind the scenes. Then we deploy and tested the model using the experiment platform to test and deploy our model. Finally when we were happy with the results we created a stremlit app to use the model and interact with the user. The app will allow users to interact with the model and see helpful insights from the data.
Challenges we ran into
It would of been helpful to have a data model for the dataset. It was hard to deploy the agent due to network issues. In addition there is not a seamless integration with agents and apps.
Accomplishments that we're proud of
We were able to create an app that was user friendly and insightful
What we learned
Its important to reduce data loads in the UI in order to provide a smooth experience for the user
What's next for Wellness Data Insights
Increase the dataset, functions and parameters given to the model. In addition, more customization for the streamlit app.
Built With
- amazon-web-services
- llama
- python-package-index
- streamlit
Log in or sign up for Devpost to join the conversation.