Ensurance

Inspiration

Selecting the right health insurance is more than just a financial decision; it’s a critical step in safeguarding one’s future. Yet, with rising costs and complex policies, many Americans feel overwhelmed and question the true value of their coverage. We envision a world where individuals aren't just consumers of healthcare, but informed decision-makers. By leveraging advanced data analytics, we strive to demystify health risks based on demographics like location, age, and gender. Our goal is to transform raw data into personalized, actionable insights, empowering users to choose insurance plans that truly align with their unique risk profiles.

What it does

Ensurance is a sophisticated data product designed to bring transparency to health risk assessment. By analyzing key demographic factors—age, gender, location, and race—the platform provides users with a comprehensive overview of potential health liabilities. Beyond risk visualization, Ensurance offers sentiment analysis to understand public perception of healthcare policies and provides curated summaries to simplify the insurance acquisition process. It’s not just an app; it’s a bridge between complex data and personal health security.

How we built it

Our development journey began with a rigorous phase of data discovery and engineering. We utilized Python and Pandas to ingest and clean diverse datasets, while Plotly powered our interactive visualizations. For real-time collaboration, our team leveraged Deepnote for notebook-based exploration and VSCode Live Share for concurrent coding. The user interface was built using Streamlit, enabling us to rapidly deploy a production-ready web application. Furthermore, we integrated Google Cloud Natural Language API for sentiment analysis, allowing us to quantify user sentiment towards the healthcare landscape.

Challenges we ran into

Data accessibility remained our most significant hurdle. Many high-resolution health datasets are gated behind paywalls or restricted access. To overcome this, we performed strategic data generalization—for instance, adapting California-specific infectious disease trends to represent a broader national context. We also navigated technical bottlenecks with external APIs; while we intended to integrate the Vericred API for direct plan recommendations, pending approval during the 36-hour sprint shifted our focus towards strengthening our core visualization and analysis engine.

Accomplishments that we're proud of

Successfully launching a functional, data-driven application from scratch in just 36 hours is a testament to our team's synergy. Despite operating across vastly different time zones, we maintained a seamless development cadence. We are particularly proud of our ability to pivot when faced with data limitations, ensuring that our core value proposition—informed health decision-making—was never compromised.

What we learned

This project was a deep dive into the intersection of data engineering and health informatics. We refined our skills in creating interactive data products with Streamlit and Plotly. On the collaborative front, we validated the efficiency of tools like Deepnote and Live Share, which proved far more effective than traditional asynchronous workflows for rapid prototyping.

What's next for Ensurance

Our vision for Ensurance includes the development of a proprietary predictive model to generate a standardized 'Healthcare Risk Score' based on granular demographic data. We also plan to integrate the Vericred API for direct insurance plan matching and expand our visualization library as more detailed datasets become available. Ultimately, we aim to offer a B2B feature where companies can utilize our sentiment analysis tools to optimize their insurance offerings based on participant feedback.

Built With

• deepnote
• gcp
• python