Ticket Platform — Infrastructure & DevOps
Scalable Cloud Setup for Real-Time Marketplace
Ticket Platform — Infrastructure & DevOps
Scalable Cloud Setup for Real-Time Marketplace
Zero
Downtime Deployments
Auto-scaling
Scraper Workers
4 Services
Containerized
The Challenge
The ticket resale platform scrapes thousands of listings, syncs prices in real time, and pushes inventory to multiple exchanges. When we first started, everything was running on a single server with no CI/CD and deployments happened over SSH. One bad deploy meant the whole platform went down and listings went stale — which directly costs money in the ticket resale business.
Our Approach
We containerized the entire application with Docker and set up an AWS infrastructure with separate services for the scraper, the API, the frontend, and a queue system for async jobs. Built a GitHub Actions pipeline that runs linting, tests, builds images, and deploys to ECS with blue-green deployment so there is zero downtime during updates. Added auto-scaling for the scraper workers so the system handles spikes during major event on-sales without manual intervention. Set up proper logging with CloudWatch and Slack alerts for anything that looks wrong.