3 real-world systems you will build end to end
You don't just watch concepts. You design, code, secure, containerize, monitor and deploy three complete platforms modelled on Amazon, Netflix and Uber.
PROJECT 1 · Modules 2–8
ShopZone: A Complete E-commerce Platform like Amazon
Build a full e-commerce backend from the first line of code to a cloud deployment. You start with 8 core services and progressively add service discovery, an API gateway, resilience patterns, event-driven communication, OAuth2 security, a full observability stack and Kubernetes orchestration on Google Cloud.
8 core services you will build
- User Service: authentication and user management
- Product Service: catalog management with Elasticsearch search
- Inventory Service: multi-warehouse stock management
- Cart Service: Redis-backed shopping cart
- Order Service: order processing with a state machine
- Payment Service: Stripe / Razorpay, UPI, wallets, COD
- Notification Service: email, SMS and push
- API Gateway: single secured entry point
What you will implement
- JWT authentication, refresh tokens, Swagger / OpenAPI docs
- Eureka service discovery and Spring Cloud Gateway
- Circuit breaker, retry, rate limiter and bulkhead with Resilience4j
- Event-driven flows with RabbitMQ, then migrated to Apache Kafka
- OAuth2, OpenID Connect and Keycloak with role-based access
- Logs, metrics and traces with Loki, Prometheus, Tempo and Grafana
- Docker, Docker Compose, Kubernetes, Helm and Google Kubernetes Engine
PROJECT 2 · Module 9
StreamFlix: A Video Streaming Platform like Netflix
Take on a media-heavy architecture: video upload, transcoding, adaptive streaming, recommendations, subscriptions and CDN delivery. This project consolidates everything from the e-commerce build under much more complex requirements.
10 microservices you will build
- User Service: OAuth2, social login, refresh token rotation
- Profile Service: multi-profile support and parental controls
- Content Service: movies and series catalog
- Video Service: upload and storage on AWS S3
- Encoding Service: FFmpeg transcoding to multiple resolutions
- Recommendation Service: collaborative, content-based and hybrid
- Subscription Service: plans, trials and billing with Stripe
- Playback Service: resume playback and continue watching
- Search Service: Elasticsearch with autocomplete and facets
- Admin Service: content approval and analytics
What you will implement
- Adaptive bitrate streaming with HLS and DASH
- CloudFront CDN with signed URLs and edge caching
- Event-driven encoding and analytics pipelines on Kafka
- Video protection with DRM, token-based access and watermarking
- Redis caching, database optimization and read replicas
- Integration testing with Testcontainers and load testing with JMeter
- Helm charts, StatefulSets, scaling and monitoring on Kubernetes
PROJECT 3 · Module 10
RideShare: A Real-time Ride-Sharing Platform like Uber
Master the hardest class of problems: real-time systems and geo-spatial data. You build live driver tracking, nearest-driver matching, surge pricing and a Kafka-driven ride lifecycle that handles thousands of concurrent users.
12 microservices you will build
- User Service: riders and drivers
- Auth Service: phone number login with OTP and JWT
- Ride Service: the complete ride lifecycle
- Matching Service: rider-driver matching algorithm
- Location Service: real-time tracking with Redis Geospatial
- Pricing Service: dynamic and surge pricing
- Payment Service: payments and wallets
- Notification Service: push notifications
- Rating Service: reviews and ratings
- Trip Service: trip history and analytics
- Driver-Onboarding Service: verification and approval
- Support Service: customer support
What you will implement
- WebSocket and STOMP for live location and ride updates
- Nearest-driver search with GEOADD and GEORADIUS
- Google Maps Directions and Distance Matrix for routes and ETA
- Surge pricing engine and demand heat maps
- Kafka event flow from ride request to payment completion
- Driver assignment optimization and fraud detection (fake GPS)
- Load testing WebSocket and Kafka throughput, then scaling on Kubernetes