Global Commodity Exchange
A real-time global commodity trading platform with a matching engine, carbon-credit listings, live WebSocket data and automated trading bots.
- Year
- 2022–25
- Role
- Full-stack engineer & DevOps: architecture to deployment
- Status
- Private · client
- Context
- Stankevicius International (via Fiverr)
Results
38+
REST endpoints
Plus 8 auth routes and a dedicated WebSocket channel
4
payment & data networks
Stripe, Binance/TRC20, SMTP, internal bots
<100ms
indexed query target
With 25–50 record payload caps
The most technically complex system I have built, and the project that moved me from websites to real-time systems. It covers automated asset listings, carbon-credit integration, a transaction matching engine, and live market data streamed over WebSockets, all developed on a dual-monitor trading desk of charts and order books.
The API surface is 38+ REST endpoints and 8 dedicated auth routes, plus a WebSocket channel for real-time data. It integrates four external networks: Stripe, Binance/TRC20, SMTP and internal loopback HTTP bots.
An automated bot-trading feature runs multi-threaded routines on cron schedules (every minute and every four minutes). Simple indexed queries target sub-100ms, with payloads capped at 25–50 records per page to keep memory flat.
I owned delivery end to end: GitHub Actions CI/CD to a DigitalOcean VPS that I set up and configured myself. Everything had to stay in sync in real time, which is where I learned to think in state, streams and latency instead of pages.