Thread by @frankc

🎉 slack.engineering/infrastructure…

A deep dive on how we crafted an order of magnitude change in our spend (10x reduction compared to baseline growth) over the last two years with iterative understanding and changes in Slack’s Continuous Integration (CI) infrastructure. 1/

👋 I referenced this article in my talk on Tracing in CI last week at
@strangeloop_stl
with excellent threads summarizing the talk by
@cyen
(

https://twitter.com/cyen/status/1444373315812945921

) /
@wiredferret
(

https://twitter.com/wiredferret/status/1444377972312530948

) 2/

I feel like theres so much MORE to explore / make in developer productivity related tooling. This body of work was a collaboration between many teams and had plenty of iterations at @slackhq and I'm excited to keep chipping away together to better understand 💯 dev / infra UX 3/

@SlackHQ In 3mon — the work on adaptive capacity internal white paper was 70 pages w/ references to @ddwoods2 / @allspaw for shaping how we understood this complex system.

@krismerrill (my manager) "this is a f'ing dissertation"

Hope this blog is actionable— plz @ me w/ questions! 4/

Oh! One thing I'd love to explore more—in a follow-up, maybe— how we used tracing with SlackTrace / @honeycombio to understand multiple hops in our CI system. It was mostly written by early employees years earlier. #observability into deployed code → safer iterative refactors 5/

@honeycombio Also if you've made it this far, would you upvote news.ycombinator.com/item?id=287883…? Appreciate you!