Cloud-cost anomaly detection
Building a cloud-cost anomaly-detection platform and the systems teams use to investigate issues.
Software Engineer / San Francisco
John Kim
Software engineer building reliable systems across infrastructure, developer tooling, and checkout.
Selected work
A few things I’ve worked on.
Less alert noise, faster resolution
Built an anomaly-resolution portal that reduced repetitive alert noise by 70%.
Checkout systems at scale
Contributed to a large-scale refactor across 600+ Sorbet T::Enum usages, checkout bug fixes, and a split-checkout frontend change associated with an 8% completion-rate increase.
Projects
What I’m building on the side.
Pocket Railway
A cozy pixel-art train game where real steps from Apple Health mine coal. Stoke the engine to travel between towns, haul freight, and furnish your carriages.
Foresight
A native journal for recording what happened, checking in on how it turned out, and reviewing patterns over time without claiming cause and effect.
Activity
A year on GitHub.
Writing
Recent notes
Junior engineers are owning more than tickets
AI agents are changing how quickly new engineers can ramp up, but ownership and judgment still belong to the person doing the work.
Read the noteKeep testing the tools you use
Why I make room to compare AI tools and agent workflows, even after finding one that works for me.
Read the note