01The problem
A monitoring service is only useful if every check runs on time and every alert arrives exactly once — for millions of websites.
02Approach
High-throughput Node.js and NestJS services run millions of website checks a day.
A queue-based alerting pipeline with retry, deduplication and SLA tracking runs on AWS ECS and Lambda.
Schema-first GraphQL APIs cut client network overhead by ~35%.
03Results
Reliable checks and alerts for 2.1M+ users.
FAQ
What did you build for UptimeRobot?
High-throughput NestJS services for website checks, a queue-based alerting pipeline on AWS ECS and Lambda, and schema-first GraphQL APIs.
Need something like this?
I'm Ahmed Mamdouh — 10+ years building systems like UptimeRobot. Scoped in one call.