Portfolio project · Video walkthroughs · English
When something breaks in the cloud at 3am, this system does the noticing, the figuring out and the explaining automatically, then waits for a human to say "yes, go ahead" before it touches anything. Watch it work, step by step.
AI Ops Serverless Platform
Technical case study · PDF
01
A three minute walkthrough of the whole system, explained over the actual architecture diagram, not slides. What breaks, how it's noticed, how AI explains it, and why a human always makes the final call. Start here if you're watching for the first time.
02
A tour through the actual code: how the entire platform is split into 8 reusable Terraform modules, and how every push to GitHub runs tests, checks security, and deploys automatically. No manual clicks in the AWS console, ever.
03
First, I check the terminal to confirm the app is healthy. Then I go to the live dashboard I built and click one button, "Break this service", to trigger a real failure. Live, no tricks. This is the moment before the AI even gets involved.
04
The moment everything comes together. The broken app gets noticed automatically, Claude explains what's wrong in Slack, and I make the final call on Mission Control, the dashboard I built myself. One click, and the service comes back to life.
Want the full technical breakdown? Read the code on GitHub →