JYY logo
← back to selected works

Digital Product · 2025

Summit AI

Simplifying complex guest and seating arrangements through AI-assisted recommendations.

Overview

Summit AI is an event management platform designed to support the planning of high-profile events involving politicians and defence diplomats. Initially developed during a hackathon and later expanded as a freelance project for MINDEF, the platform helps organisers manage guest lists, attendance and protocol-sensitive seating arrangements. Its AI-assisted replacement workflow recommends suitable delegates when seating changes occur, while allowing organisers to review suggestions, understand related changes and retain manual control.

Background

Summit AI began as a hackathon challenge focused on simplifying the complex process of seating delegates at the Shangri-La Dialogue. Each foreign dignitary has a specific level of precedence, along with seating rules and protocol requirements that organisers must follow. The process is made even more challenging by last-minute changes, such as cancellations, replacements and changes in attendance.

The challenge was provided by the organisers of the Shangri-La Dialogue as part of an internal, interdepartmental hackathon organised by MINDEF, with participation from DSTA.

Our team was assigned this problem statement and developed a minimum viable product over the course of one week. We used Streamlit to build the initial interface and demonstrate how the proposed solution could support event organisers.

My Role

After the hackathon, Summit AI was approved for further development, and I was engaged as a freelance designer for three months.

As the main designer, I was responsible for transforming the early MVP into a polished, launch-ready product built with React and Material UI. This involved refining the user experience, improving the interface and developing the original proof of concept into a more complete and practical system.

I also created a distinct visual identity for Summit AI and developed a centralised design system to maintain consistency across the product and my other MINDEF projects.

Research

The research process began with understanding the protocol and seating rules used at the Shangri-La Dialogue. As the platform needed to handle large amounts of guest information, we explored ways to present complex details clearly without overwhelming organisers.

We studied table management tools, wedding planning platforms and event seating applications to understand how they handled guest lists, seating layouts and last-minute changes. We also reviewed data dashboards and scheduling tools to learn how filters, search functions and expandable sections could help users find important information quickly.

These findings guided us towards a system that offered clear visualisations, flexible table management and detailed drill-down options. Organisers could view the overall seating plan at a glance while still accessing the information and rules linked to each delegate when required.

Design and Iteration

By creating high fidelity wireframes to explore the structure of the platform and map out its key user flows. These early designs focused on organising complex information clearly, simplifying table management and helping organisers respond quickly to changes.

The wireframes were then developed into an interactive prototype, allowing stakeholders to experience the proposed workflows and provide more detailed feedback.

Weekly discussions with the cross-functional project team helped us align on project goals, manage priorities and continuously improve the design. Feedback gathered during these sessions shaped each iteration, resulting in a more practical, intuitive and user-friendly product.