All thoughts
Dev Notes1 min read
How We Cut Infra Costs by 40% (Without Breaking Anything)
No magic. Just killing what shouldn't have been running in the first place, then building habits so it doesn't creep back up.
No magic, just unglamorous cleanup
Cutting 40% off our AWS bill sounds like some genius optimization move. Honestly, it was mostly janitorial work, finding stuff that was quietly running and costing money for no reason.
- Killed unused EC2 instances left over from old client projects
- Cleaned up orphaned EBS volumes nobody remembered creating
- Right sized instances that were massively over provisioned
- Moved a few workloads onto Redis caching instead of hitting the DB every time
The bigger win wasn't the one time cleanup, it was building the habit of a monthly cost review so waste doesn't quietly creep back in. 35+ live apps means small leaks add up fast if nobody's watching.
Audit every running resource against an actual owner
Set up billing alerts before costs spike, not after
Make cost awareness part of the team's culture, not just my job
The cheapest server is the one you remembered to turn off.
It's not a flashy story, but it's the kind of unglamorous engineering work that actually moves the needle when you're managing infrastructure at this scale.
2 views
Comments (0)
Loading comments...