Carebricks

• 

  Our Logo

Inspiration

Several of us have firsthand experience with how the Indian healthcare system works, especially in rural areas where the gaps in care are widest and the hardest to see. We wanted to give the people who plan that care a clearer way to find those gaps and act on them.

What it does

CareBricks pairs AI and traditional statistics with a simple interface for health workers and planners. It shows where care is missing, separates real gaps from places we just don't have enough data on, and turns that into a concrete plan of action.

How we built it

We built CareBricks on Databricks using the Lakehouse and Lakebase for data storage. We used Lakeflow pipelines to move data through a medallion architecture and Databricks Genie for natural-language querying. The frontend is built with React.js, and OpenAI APIs generate intervention plans for regions identified as medical deserts.

Challenges we ran into:

One of our biggest challenges was integrating our Medical Desert Planner with Lakebase. Despite guidance from mentors at the Hackers Corner, we weren't able to fully complete the integration during the hackathon. We adapted our architecture to keep the project functional and focused on delivering a working solution.

Accomplishments that we're proud of:

We built a working end-to-end application during Databricks Summit! One of our team members is a student training to become a data engineer so it was a great learning experience for them to learn about combining AI, analytics, and data engineering into a single workflow works. It was great accomplishment to design a system that accounts for uncertainty instead of treating missing data as a confirmed healthcare gap.

What we learned

We learned how to build and deploy a complete data pipeline on Databricks using a medallion architecture, and how important it is to model uncertainty responsibly when working with real-world healthcare data.

What's next for Carebricks

With more time and resources, we can integrate richer and more reliable datasets on Indian healthcare infrastructure. We can expand coverage of hospitals, clinics, pharmacies, and community health centers, incorporating population health and demographic indicators. A great outcome would be providing policymakers with scenario-planning tools to evaluate the impact of future investments before resources are deployed.

Built With

• databricks
• genie
• lakeflow
• python
• sql