All work

Real-time systems

CryptoTracker

A containerized BTC, ETH, and SOL tracker with live WebSocket streaming, historical persistence, security controls, and Kubernetes manifests.

SYSTEMReal-time market signal pipeline
EVIDENCEPublic repository · verified
SOURCEGitHub repository
ARCHITECTURE / CONCEPTUAL VIEW

Derived from public repository documentation

01Binance Streaminfrastructure
02Node.js APIservice
03WebSocket Clientinterface
04PostgreSQLdata
05Kubernetesinfrastructure
01

CHALLENGE

What the system needed to solve.

Live market data is only useful when the experience stays current, reconnects cleanly, persists history, and protects its application surface.

02

CONSTRAINTS

The boundaries that shaped the build.

  • Stream changing prices without constant page refreshes.
  • Persist enough history for useful trend views.
  • Package the system for local and production-like deployment.
03

ENGINEERING DECISIONS

Choices with consequences.

Streaming and history have separate paths

WebSockets carry live updates while PostgreSQL backs historical queries, keeping each interaction aligned with its data shape.

Failure recovery is visible

The client reconnects after stream interruption rather than assuming a permanent connection.

Deployment is documented as code

Docker Compose and Kubernetes manifests describe the application, API, database, ingress, and secret boundaries.

04

RELIABILITY CONSIDERATIONS

Prepared for the operating reality.

  • Origin validation and rate limiting
  • Indexed historical price queries
  • Health endpoint and reconnect behavior
05

OUTCOMES / HONEST EVIDENCE

What can be stated today.

  • The public repository documents the full stream-to-storage architecture.
  • Load-test results and production traffic metrics are pending.

TECHNOLOGY

  • Next.js
  • Node.js
  • WebSockets
  • PostgreSQL
  • Kubernetes