Full-Stack Developer & AI Engineer · Intern @ Tata Technologies
Processes business files in under 90 seconds — from upload to cited AI insights, live in production
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 their position. Receptionists juggle paper and verbal announcements. Doctors have no dashboard. Built for Queue Cure '26 Hackathon by Wooble × Unstop, with the core requirement: both receptionist and patient screens must update instantly when "Call Next" is clicked, with zero refresh. Process I chose Socket.io over polling because the requirement demanded sub-second updates across multiple screens. I designed the wait-time calculation as position × avgConsultationTime, computed live and never hardcoded — receptionists can adjust consultation time at any moment and all wait times recalculate instantly across every connected screen. For the receptionist UI, I prioritized mistake-proofing: confirmation dialogs for destructive actions, disabled states when the clinic is closed, and toast feedback on every action. I separated "Skip" from "Remove" as distinct actions to prevent accidental data loss. I documented concurrency explicitly — Node.js processes Socket.io events on a single thread, so multiple receptionists clicking "Call Next" simultaneously cannot race condition; events Results Live sync confirmed at under 50ms across all connected screens in production testing. Successfully deployed full-stack: backend on HuggingFace Spaces (Docker), frontend on Netlify, with GitHub auto-deploy CI/CD. Built and shipped a working AI chatbot, voice announcement system, and live analytics dashboard beyond the hackathon's core requirements — demonstrating both speed and depth under a 9-day timeline. Reflection I'd add MongoDB persistence so queue data survives server restarts — currently it's in-memory, which resets if the HuggingFace Space restarts. I'd also add SMS notifications via Twilio so patients don't need to keep the browser tab open to know when they're called.
This is Mohit’s work on Wooble.
Build a profile that shows what you can do — and share it anywhere.
