📊 Real-Time Analytics Platform

A production-style real-time analytics platform that collects user events from websites/apps, processes them through a queue system, stores structured analytics in PostgreSQL, caches frequently accessed queries in Redis, and powers a live dashboard UI with WebSockets.

Built to simulate how large-scale analytics platforms like Mixpanel, PostHog, or Google Analytics work internally.

---

🚀 Features

• Event Ingestion API
  • Accepts events like PAGE_VIEW, BUTTON_CLICK, SESSION_START, PURCHASE
• High Performance Backend
  • Node.js + Express + TypeScript
• Reliable Storage Layer
  • PostgreSQL for structured analytics data
• ⚡ Redis Caching
  • Cache-aside strategy to speed up analytics queries
• 📬 Queue System
  • Handles burst traffic and async processing
• 📡 Real-time Updates
  • WebSocket live metrics streaming
• 📈 Analytics Dashboard UI
  • Charts, metrics, top pages, recent events
• Docker Ready
  • Run entire stack with one command
• Production-like Architecture
  • Logging, rate limiting, environment handling

---

🏗️ Architecture

Client / App / Website | v /api/events ---> Express + TypeScript | Queue (Bull / Redis) | v PostgreSQL (Events DB) | Analytics Services | +--> Redis Cache | +--> REST Analytics API | +--> WebSocket Live Stream | v React Dashboard UI

---

🛠️ Tech Stack

Backend

• Node.js
• Express
• TypeScript
• PostgreSQL
• Redis
• Queue (Bull / RabbitMQ)
• WebSockets

Frontend

• React
• Vite
• Chart.js
• TailwindCSS

DevOps

• Docker + Docker Compose
• Prometheus + Grafana (optional monitoring)

---

🧪 API Endpoints

📥 Event Ingestion

Body

{
  "event_type": "PAGE_VIEW",
  "user_id": "user_101",
  "session_id": "sess_1",
  "page_url": "/pricing",
  "metadata": {
    "device": "mobile",
    "browser": "Chrome"
  }
}


GET /analytics/summary
GET /analytics/top-pages
GET /analytics/timeseries?range=24h
GET /analytics/recent
▶️ Running Locally
Backend
npm install
npm run dev

Dashboard
cd dashboard
npm install
npm run dev


Open browser at:

http://localhost:5173

🐳 Run Entire System with Docker
docker-compose up


This will start:

Backend

PostgreSQL

Redis

Monitoring stack (optional)

🧼 (Optional) Seed Dummy Data

To reset:

TRUNCATE TABLE events RESTART IDENTITY;


Then insert dummy dataset or auto-generate test events.

📸 Screenshots

Add screenshots in repo:

/screenshots/dashboard.png
/screenshots/timeseries.png
/screenshots/top-pages.png

🧠 What I Learned

Designed a scalable backend architecture similar to production analytics platforms

Implemented Redis caching strategies (cache-aside, TTL)

Built queue-based async processing for handling burst traffic

Designed REST analytics APIs and time-series analytics

Implemented real-time WebSocket streaming

Structured backend using modular & layered architecture

Built a complete analytics dashboard UI using React + Chart.js

🚀 Deployment

Planned deployment targets:

Backend: Render / Fly.io / VPS

Frontend: Vercel

Database: Managed PostgreSQL

Redis: Upstash / Render Redis

💡 Possible Enhancements

Authentication + multi-tenant analytics

Public JavaScript SDK (auto tracking)

Custom dashboards per app

Funnel tracking

Alerts & anomaly detection

More event schema analytics