SAFIYA BEGAM K

Full-Stack Developer

Dhanalakshmi Srinivasan Engineering Collegefull_time, internship, freelance

Open to roles

GITProblem SolvingCommunicationAgile/Scrum

GitHub LinkedIn Personal website· safiya-portfo.vercel.app

EXPIRY ALERT ' 26

How to Run Locally Prerequisites Node.js v18+ A MongoDB database — local mongod, or a free MongoDB Atlas cluster A Gmail account + App Password (for email alerts/import — optional, app works without it) Step 1 — Install everything npm run install:all Step 2 — Configure environment Copy server/.env.example → server/.env (or edit the existing .env) and fill in: MONGODB_URI=your-connection-string EMAIL_USER=youraddress@gmail.com EMAIL_APP_PASSWORD=your16charapppassword EMAIL_ALERTS_ENABLED=true For email import, also enable IMAP in Gmail: Settings → Forwarding and POP/IMAP → Enable IMAP. Step 3 — Seed demo data (optional but recommended) npm run seed Step 4 — Run npm run dev Backend → http://localhost:5000 Frontend → http://localhost:3000 ☁️ Deployment Deployed as two separate services: Backend (Express + Socket.IO) → Render, root directory server Frontend (React) → Vercel, root directory client Database → MongoDB Atlas The frontend reads two build-time environment variables so it can be deployed independently of the backend's domain: REACT_APP_API_URL=https://your-backend.onrender.com/api REACT_APP_SOCKET_URL=https://your-backend.onrender.com The backend's CLIENT_URL env var must point back at the deployed frontend domain, for CORS. Note: on Render's free tier, outbound SMTP/IMAP connections may be restricted, so email alerts/import are most reliable when run locally or on a paid tier. 🔌 API Reference Method Endpoint Description GET /api/records List records (?status=, ?category=, ?search=) GET /api/records/stats Dashboard stats + upcoming expiries POST /api/records Create record PUT /api/records/:id Update record (triggers reclassification) DELETE /api/records/:id Archive record (soft delete) POST /api/records/recalculate/all Manually recalculate all statuses POST /api/records/:id/send-alert Manually (re)send the expiry alert email POST /api/records/:id/attachment Upload/replace a record's document GET /api/records/:id/attachment Download the attached document DELETE /api/records/:id/attachment Remove the attached document GET /api/activity Recent activity log entries GET /api/email-import/status Whether email import is configured POST /api/email-import/scan Scan inbox, auto-classify, and create records Socket.IO Events Event Direction Payload record:created / record:updated Server → Client { record, stats, origin } record:deleted Server → Client { recordId, stats, origin } records:recalculated Server → Client { stats } activity:new Server → Client { entry } presence:update Server → Client { count } origin carries the originating client's socket id, so a client can tell "my own action" apart from "someone else's" and only toast for the latter.

4 media files · record-vault-five.vercel.appView

Queue Cure — Digital Clinic Queue Manager

# 🏥 Queue Cure — Digital Clinic Queue Manager 76% of India’s 1.5M clinics still run on paper slips and shouting. Patients wait 2–3 hours with zero visibility. Doctors have no dashboard. Receptionists manage everything from memory. **Queue Cure** fixes this with a live digital queue manager — visible to patients on their phone, controlled by the receptionist on one screen. --- ## ✨ Key Outcomes - **Receptionist efficiency:** Add patient + assign token in **<10 seconds** - **Patient transparency:** Live updates on phone — **no refresh needed** - **Real data:** Estimated wait time computed Process We started by framing the problem with evidence: 76% of India’s clinics still rely on paper slips and shouting. Patients wait 2–3 hours with no visibility, doctors lack dashboards, and receptionists manage everything from memory. Our approach was lean and practical: Researched clinic workflows by observing reception desks and patient waiting rooms. Identified three critical requirements: add patient in <10s, live updates without refresh, and wait time from real data. Chose a lightweight tech stack (Node.js, Express, Socket.IO) for real‑time sync. Iterated quickly: first tried polling with setInterval (too slow), then switched to WebSockets for instant updates. Designed two screens: Receptionist dashboard (add patient, call next, set avg time) and Patient view Results Receptionist can add patient and assign token in <10 seconds. Patient screen updates instantly when “Call Next” is clicked — no refresh required. Estimated wait time is computed from queue length × avg consultation time, not hardcoded. Demo moment: “The moment the receptionist clicks Call Next, every patient’s phone updates instantly with their new wait time — no shouting, no confusion.” Feedback from early testers: patients felt less anxious knowing their wait time, receptionists reported reduced chaos. Reflection Add SMS/WhatsApp alerts so patients don’t need to stay in the waiting room. Build a doctor dashboard to track consultation stats and throughput. Include analytics for clinic owners (average wait, patient flow, busiest hours). Improve accessibility with multilingual support for rural clinics. Next iteration: integrate with hospital management systems for seamless adoption.

View

This is SAFIYA’s work on Wooble.

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

Build yours