The Story Behind Our Insurance Tracking Application

Inspiration

The idea for our insurance tracking application came from a personal experience. I was planning to get some routine health tests done and had seen on social media that, in some cases, these could be covered by insurance. I wanted to check if my insurance covered them too.

However, when I tried to find out, it turned out to be a nightmare. Policy documents were filled with tiny, unreadable text and legal jargon that made it almost impossible to understand what was actually covered. This made me realize how difficult it is for most people to truly understand their insurance policies.

Adding to this, I was already aware that insurance claim disputes are quite common in India. Many end up in court simply because of a mismatch between what people think their policy covers and what the insurer is willing to offer. A lot of this stems from poor communication or misleading information from agents.

That's when the idea clicked why not build an app that helps people manage all their insurance policies (health, car, life) in one place, and most importantly, helps them actually understand what's written in their policies in simple, jargon-free language?

How We Built It

I teamed up with a friend to bring this idea to life. We started by building a quick prototype using Python and Streamlit. This helped us validate the core idea and user experience.

Once we had a working version, we moved on to building the full application using Next.js for the frontend and NeonDB for the backend. We initially explored building a RAG (Retrieval-Augmented Generation) system using open-source models, but later decided to integrate with the Perplexity API because of its strong search capabilities, which significantly improved the accuracy and relevance of policy information retrieval.

Challenges Faced

One of the biggest challenges we encountered was the lack of reliable and structured information around insurance policies. This made it hard to decide which features to prioritize and how to present policy details in a meaningful way. We had to spend a lot of time researching how insurance typically works in India and identifying the pain points users face.

What We’re Proud Of

We’re proud to say that we’ve been able to solve the core problem to a large extent. With the help of this application, users can directly ask questions and get answers based on their policy documents — without having to struggle through endless pages of tiny-font legal text.

The Road Ahead

Looking ahead, we hope to collaborate with insurance providers. Such partnerships would give us access to more detailed customer and policy data, helping us further improve the accuracy and usefulness of our application.

With these collaborations, we aim to expand our platform to let users pay premiums, buy or modify insurance policies, and ultimately provide a 24x7 insurance support agent because accidents and emergencies don’t wait for business hours.