SmartWait

Inspiration

The inspiration for SmartWait comes from a shared, universal experience: the anxiety and frustration of sitting in a crowded healthcare waiting room. The lack of information, the uncertainty of wait times, and the operational inefficiencies for staff create a stressful environment for everyone. I wanted to build a solution that brings modern, real-time technology to this outdated process, giving patients back their time and providing healthcare staff with the tools to manage patient flow effortlessly.

What it does

SmartWait is a full-stack, virtual queue management system.

• For Patients: Patients can check in via a mobile-friendly web portal (e.g., by scanning a QR code). They immediately receive their queue position and an estimated wait time via SMS and a live status screen. They are then free to wait anywhere, receiving automated SMS alerts when it's time to get ready and when it's their turn to be seen.
• For Staff: A secure, real-time staff dashboard provides a complete overview of the patient queue. With single-click actions, staff can "call" the next patient or mark a visit as "complete." The system instantly updates the queue for everyone and triggers the necessary patient notifications, eliminating manual coordination and reducing administrative workload.

How we built it

SmartWait was built from the ground up in a unique partnership with Kiro, my AI collaborator. The process was a phased journey, moving from high-level strategy to focused execution.

1. AI as Architect: The project began not with code, but with planning. Using Kiro's brainstorming and spec-to-code capabilities, we created a comprehensive project blueprint inside the /.kiro directory. This included a full Product Requirements Document, technical designs, and even steering documents for API standards and healthcare compliance. Kiro helped architect a professional-grade system before a single line of application code was written.

2. AI as Agile Partner: The initial vision was vast. Realizing the constraints of the hackathon, I worked with Kiro to pivot. We collaboratively generated a new spec, a focused 10-day MVP plan, which became our guide. This demonstrated a powerful workflow where the AI partner helped adapt the strategy to meet real-world deadlines.

3. AI as Pair Programmer: During the build phase, Kiro's inline AI coding was indispensable. Adopting a Test-Driven Development (TDD) approach, I would write the unit tests for a service, and then prompt Kiro to generate the business logic to make them pass. Together, we implemented the Prisma ORM, built the complex WebSocket service with reconnection logic, integrated the Twilio API for SMS, and developed the full-stack staff authentication system. Kiro accelerated development dramatically, while also helping refactor and debug along the way, ensuring the final code was robust and maintainable.

Challenges we ran into

The primary challenge was scope. My initial vision was an enterprise-level system, which was far too ambitious for a hackathon. The key was not to abandon the vision but to distill it. Collaborating with Kiro to create the focused MVP task list was the solution; it turned an overwhelming project into a series of achievable daily goals.

Technically, implementing a robust real-time WebSocket service with reliable reconnection logic and state management was a significant hurdle. This is where the pair-programming dynamic with Kiro shone. By breaking the problem down and using Kiro to generate boilerplate for connection management, event handling, and custom React hooks, I was able to build a production-quality feature in a fraction of the time it would have taken alone.

Accomplishments that we're proud of

First, I'm proud of building a complete, end-to-end, production-ready application, not just a proof-of-concept. It has a secure data layer, a real-time communication channel, third-party API integrations, and distinct user-facing interfaces for both patients and staff.

Second, I'm proud of the development process itself. By partnering with Kiro, I was able to adhere to professional best practices like Test-Driven Development, creating a well-architected monorepo, and thoroughly documenting the deployment process.

Finally, I'm proud of the story this project tells: that a single developer, partnered with an AI, can move from a big idea to a deployed, functional product with incredible speed and quality.

What we learned

The biggest takeaway is that AI in development is far more than an auto-complete tool—it's a genuine collaborator. Kiro wasn't just a code generator; it was a partner in product definition, architectural design, strategic planning, and agile execution. This project taught me a new workflow where I, the developer, set the direction, architecture, and quality bar (by writing tests), while the AI partner helps with the heavy lifting of implementation, allowing me to stay focused on the bigger picture.

What's next for SmartWait

The solid foundation we built during the hackathon is just the beginning. The original vision we architected with Kiro is our roadmap. The next steps are to build out the features we planned from day one:

• Advanced Analytics Dashboard: Providing healthcare facilities with insights into wait times, staff performance, and patient flow.
• EHR/PMS Integration: Connecting SmartWait directly with existing healthcare record systems using the HL7 FHIR standards we defined.
• ML-Powered Wait Time Predictions: Moving beyond simple estimates to a machine learning model that predicts wait times with greater accuracy based on historical data.

Built with

• Backend: Node.js, Express, TypeScript, Prisma ORM, Socket.io
• Frontend (Web Portal & Staff Dashboard): Next.js 13, React, Tailwind CSS
• Mobile (Patient App): React Native, Expo
• Databases: PostgreSQL, Redis (for caching and WebSocket scaling)
• Infrastructure & Deployment: Docker, Docker Compose
• APIs & Services: Twilio API (for SMS notifications)
• Development Tools: Jest (for testing), npm Workspaces (for monorepo management)

Built With

• docker
• express.js
• nextjs
• node.js
• postgresql
• prisma
• react
• reactnative
• redis
• socket.io
• tailwindcss
• twilio
• typescript