Roopesh G A

Full-Stack Developer

Vellore Institute of Technologyfull_time, internship, freelance

Open to roles

CommunicationProblem SolvingProject ManagementLeadership

MemoryVerse AI '26

By clicking the Sign in, you can use these test credentials Email: demo@memoryverse.ai Password: MemoryVerse2026

5 media files · memory-verse-ai-sepia.vercel.appView

Queue Cure '26 — Real-Time Clinic Queue System --- https://github.com/Roopesh2006/Queue-Cure

76% of India's 1.5 million clinics still run on paper token slips and shouting. Patients wait 2-3 hours with zero visibility into when they'll be called. Existing solutions like Practo Ray and SmartClinic bundle queue management inside 30+ module suites requiring onboarding and staff training — completely unusable for a single-doctor walk-in clinic. No free tool offered a mobile-first, no-install patient waiting room with dynamic wait estimates. I chose to build the smallest possible system that solves exactly this gap: two URLs, one for the desk, one for patients. Process I started by mapping the two personas — the receptionist who manages 40+ daily walk-ins from memory, and the patient with a paper token and no idea how long they'll wait. I chose Socket.io over HTTP polling because the 40% evaluation weight on live updates required zero-refresh sync. For wait time, I built a rolling average of the last 5 actual consultation intervals instead of a hardcoded number — this directly addressed the 25% rubric weight on real data. I used Upstash Redis for persistence so the queue survives server restarts. The receptionist PIN protection prevents accidental "Call Next" from a patient who navigates to the wrong URL. Results Deployed live on Render with both receptionist and patient screens fully functional. Queue updates propagate to all connected screens in under 200ms. Wait estimates self-calibrate after each Call Next — if the doctor averages 3 minutes today, the system learns this and corrects estimates in real time. Patients need only a browser and a URL — zero app install. The system handles multiple simultaneous connections with Socket.io's built-in reconnect and Redis-backed state. Reflection I'd add the redirect rule on Render before deploying so the patient waiting room URL works on direct access without extra configuration. I'd also seed the queue with test data automatically on first load to make the judge demo smoother. Post-hackathon, I'd add WhatsApp notification via Meta Business API to alert patients when 2 tokens away — eliminating the need to keep the browser open entirely.

2 media files · https://queue-cure-1-ihym.onrender.com PIN-1234View

This is Roopesh’s work on Wooble.

Build a profile that shows what you can do — and share it anywhere.

Build yours