· 1 min read
Batch size, two years after DORA measured the AI paradox
Two years after DORA quantified the AI paradox, the pattern in our own reviews has resolved into something specific enough to act on.
#batch-size#ai#dora#delivery
Delivery
9 articles
· 1 min read
Ask the queue, not the team
A one-minute note on the first question of any delivery diagnosis.
#queues#diagnosis
· 1 min read
Code review is the queue nobody staffs
Google's engineering guidance treats review speed as a first-class value. Most organizations treat it as weather. The gap is a finding.
#code-review#queues#delivery
· 1 min read
Trunk-based development: the evidence and the tantrums
The research keeps finding the same thing, teams keep arguing anyway, and the argument usually is not about branching. What the fight is actually over.
#trunk-based#delivery#dora#batch-size
· 1 min read
Little's law sets your lead time, not your roadmap
Work in progress, throughput, and time in system are one equation. Your delivery numbers already obey it, whether or not anyone is looking.
#queues#wip#delivery
· 2 min read
A delivery overhaul, read as a worked example
One public writeup of a real release-process rebuild, and the three transferable decisions inside it that had nothing to do with tooling.
#delivery#release-process#case-study
· 2 min read
Platform engineering pays rent in queue time
The research says platforms help and hurt on different clocks. A guide to reading your own platform investment straight, at month three and month eighteen.
#platform-engineering#dora#delivery
· 1 min read
Deploy frequency is a symptom, not a lever
The number is worth tracking and useless to push on. What moves it lives upstream, in queues nobody owns.
#delivery#queues#metrics
· 1 min read
AI made coding faster and delivery slower
The 2024 DORA report measured the paradox precisely. The mechanism is batch size, and it was in the data before AI arrived.
#dora#ai#batch-size#delivery