How Our Team Built an Insurance Assistance Application with PartyRock

Tehhp
Government Digital Products, Singapore
3 min readMay 10, 2024

This year’s AWS Summit introduced an exciting new hands-on session called the AWS Generative AI Hackathon with PartyRock. Intrigued by the possibilities of Generative AI, my colleagues and I eagerly decided to participate in the event to gain a deeper understanding of how GenAI could revolutionize application development without the need for traditional coding. The event commenced with a brief introduction by the PartyRock team, outlining the workings of PartyRock and its innovative approach to application creation.

PartyRock, a playground powered by AWS Bedrock.
AWS Generative AI Hackathon with PartyRock event

Identifying the Need

Following the introduction, the host briefed all participants on the tasks, rules, and judging criteria for the hackathon. It was clear that to succeed in this challenge, we needed to address a significant issue: the difficulty insurance sales agents face in managing policy details and customer needs.

The Approach

Armed with our insights, we adopted a prompt engineering approach to list down our requirements in PartyRock. We clearly defined what inputs are and what the expected outputs should be. Our goal was simple yet ambitious: to create an application that would notify agents six months prior to a policy expiration and suggest new policies based on existing customer details.

Building the Application

With our requirements in hand, we unleashed the power of PartyRock to bring our vision to life. The application was generated effortlessly, thanks to the intuitive nature of this playground. However, our journey didn’t end there.

Testing and Refinement

We understood the importance of testing, so we tested each function of the application to ensure it performance the expected. We also took the opportunity to refine the application, making it more suitable for our specific use case. One key enhancement was the addition of a chatbot widget, which provided sales agents with valuable tips on approaching customers about new policies.

Final Touches and Pitch

After multiple rounds of testing and refinement, we were confident in the capabilities of our application. We made the final touches and prepared a compelling 3-minute pitch to present our solution to the judges. Our hard work and dedication paid off when we were awarded 3rd place in the hackathon.

Small token but big encouragement. Cheers to Alvin Siew, Zhi Xuan, Zeng Bo, Yuli & me. :)

Conclusion

Our experience in developing an insurance assistance application with PartyRock underscored the impact of innovation and collaboration. Leveraging this cutting-edge technology, we successfully created a prototype application that addresses a critical need in the insurance industry, empowering sales teams to excel. This journey serves as a testament to the potential of advanced tools like PartyRock and reinforces the notion that with the right tools and mindset, any challenge can be overcome.

PolicyPro — https://partyrock.aws/u/tehhp/uyeFHYh7U/PolicyPro

--

--