Building better standups

Dog Days of DevOps 2025

Full remote (though a bunch of us are in Austin)
Open-core (and a lot of our processes are public)
Sprints every 3 weeks
Minor releases every ~3 weeks, patches as needed
3 engineering teams, each with
- 4-5 backend devs (varying degrees of full-stack) <- I am here
- 1 each of primarily-FE dev, EM, product designer, QA
I'm on the software team, which includes vuln management
- Some of our workload gets delivered outside release cycles
- Rituals are a little different team-to-team

Async what you can
Unblock who you can
Meetings should have agendas before and records after
- Calls are generally recorded, including standups, which helps async-ness
Focus on finishing (start at the right side of the board)
Focus on the work rather than the people, except when unblocking people
Don't let ritual work pile up
Work in descending order of how many people need to be on the call

Async standup init'd via Geekbot
- Pings folks when they come online in the AM
- Traditional questions
  - What did you do since last update
  - What will you do between now and next update
  - Blockers?
- Dumped into team Slack; only blockers are discussed in the meeting
Daily synchronous standup agenda (GDocs tabs FTW)
- "Agenda" because folks can add items ahead of time
- Meetings tend to run 30m...feels long but the ritual carries a lot of weight
For the rest of this presentation, let's look at the doc

Questions? Find me here / @ian@phpc.social / @ian.im / @iansltx

By Ian Littman