The Kerf blog
Notes on statistics, engineering and product practice from the people who build Kerf.
Latest

Guardrails, not gates
Why the best programs automate safety checks instead of adding approval steps.

The winner's curse in A/B testing
Why the lift you measured is larger than the lift you will get.

Feature flags are debt, so pay it down
Ownership, expiry and cleanup for a healthy flag system.

An AI wrote our experiment readout
What the model does well, where it needs limits and how we check it.

Holdouts at scale
A long running holdout answers whether your program is working.