About the challenge
Build an AI agent, agent system, or agent application using the Databricks platform. While you're free to work on any project that excites you, we encourage exploring our "data for good" themes using our partner datasets (details below). The only requirement is using Databricks as your platform, though you can integrate any additional tools or frameworks you'd like. Some things to consider:
- You have approximately 6 hours, so think about what you can realistically accomplish and do some planning in advance
- You'll be giving brief presentations to judges, so consider how to develop a minimum viable product (MVP) that will demo well
- Explore the partner sites/resources beforehand to understand how you might be able to integrate them into your projects
Get started
- Sign up for your Databricks trial using this link, which will give you $1,000 in credits to use for the hackathon for the day.
- Get access to partner data by submitting your workspace's Delta Sharing ID via this form. Instructions for obtaining your workspace's Delta Sharing ID and setting the required permissions can be found here.
- Read through some of the notebooks and other materials in this GitHub repo for ideas and inspiration.
Data for Good Challenges
We've partnered with organizations providing datasets to support projects with social impact. While working on these challenges and using partner data is optional, we highly recommend it—it's a great opportunity to make an impact and connect with some great organizations.
Theme 1: Community Wellness & Support Navigator AI agents that provide insights into community health, connect people to local support services (health, social, wellness), and help them navigate complicated healthcare benefits data.
Theme 2: Accessible City & Travel Equity Agent AI agents that promote inclusivity by helping individuals with accessibility needs discover and navigate cities and travel options, or analyze urban data to advocate for improvements.
Data partners: The Bright Initiative (Google Maps business data, Airbnb and Booking.com datasets), mimilabs (30TB+ healthcare data from CMS, CDC, FDA), and Nimble (live web data from Google Maps places, reviews, product listings).
Follow this link for instructions on accessing partner data in your workspace, and see this notebook for a quickstart guide on using the partner data once it is in your workspace.
Location:
888 Howard St, San Francisco, CA 94103, USA
