← Back
Cloud project · AWS · Learning build

AWSRide-Share

A production-grade ride-sharing app built from scratch to master AWS serverless architecture — Amplify, Cognito, API Gateway, Lambda, DynamoDB, all wired together.

0

AWS services

0%

serverless

0 servers

to manage

AWS ride-share - demo
Why I built this

AWS has 200+ services.
I needed to actually use them.

The problem

  • Docs and tutorials only go so far
  • No feel for how services connect
  • Never built a real-time serverless app
  • IAM, Cognito, WebSockets — all abstract
  • Needed something end-to-end to make it click

The approach

  • Build a full ride-sharing app from scratch
  • Wire up every core AWS service manually
  • Implement real auth, real APIs, real DB
  • Deploy with proper CI/CD from day one
  • Learn by breaking things in production
Architecture

100% serverless,
end to end.

Request flow

UserAmplify (frontend + CI/CD)
AuthCognito (JWT tokens)
APIAPI Gateway (REST + WebSocket)
LogicLambda (ride matching)
DataDynamoDB (rides, users, locations)

Ride flow

01User authenticates via Cognito
02Ride request hits API Gateway
03Lambda matches rider with driver
04Real-time updates via WebSocket
05Ride data persisted to DynamoDB
AWS services

Every service, hands-on.

Click any card to see how and why it was used.

What I learned

6 services.
One working app.

Serverless architecture patterns
Real-time WebSocket implementation
NoSQL data modelling for ride queries
IAM least-privilege security
Cognito auth flows end-to-end
API Gateway REST + WebSocket config
CI/CD pipeline with Amplify
Event-driven Lambda functions
Stack
Frontend
AWS AmplifyGitHub CI/CDHosted UI
Auth
Amazon CognitoUser PoolsJWT tokens
API
API GatewayREST APIWebSocket API
Logic
AWS LambdaNode.jsRide matching
Database
DynamoDBGSI indexesMillisecond reads
Security
IAM rolesLeast-privilege policiesService-to-service auth