Snapshots

We deploy and sweep constantly, yet no weekly meeting exists

Waiting on you
  • Every tested change goes straight to staging, where Param tests on the live URL.
  • QV rounds follow every phase, and QA sweeps start when Sai's sheet fills.
  • The handbook runs at 9 am; brands get a Monday 09:00 digest.
  • A restore test runs monthly; the first was on 25 Oct.
  • Proposed: a 30 minute Monday review covering Next, partners and decisions.
49bugs closed, one QA sweep
Your call

Hold a weekly 30 minute review on Mondays at 11 am IST from 2 Nov?

People and ways of working · 14.3

Rituals

DraftLast checked Mon Oct 26Owner Param

Answer

The team's only fixed rhythms today are continuous deploys, QV rounds after every phase, QA sweeps from Sai's sheet and the handbook's 9 am run; there is no weekly meeting written down, so this page records what happens and proposes a weekly review.

Today

Ritual When What happens Source
Continuous deploy After every tested change Deploy to staging; Param tests on the live URL brew/CLAUDE.md memory always-deploy-to-test
QV round After every phase of phased work, then 3 more before closing Fresh reviewers, live checks on staging, fixes with tests, redeploy, re-verify memory brew-qv-process
QA sweep When Sai's QA sheet fills Every row triaged, fixed, then a compare pass; 19 to 20 Oct closed 49 bugs memory brew-qa-sweep-2026-10-19
Design partner audit 20 Oct and 24 Oct so far About 86 findings on 20 Oct; report A, B and C on 24 Oct memory brew-qa-audit-2026-10-20 memory brew-qa-round2-2026-10-24
Restore test Monthly; first on 25 Oct Production export restored and counted brew/.wt-prod/docs/production.md:43
Handbook morning run 9 am Freshness and truth checks, overnight commits, Next, a short digest ~/.claude/scheduled-tasks/handbook-morning/SKILL.md
Weekly digest to brands Monday 09:00 brand time The product emails each team what changed brew/digest.js:1-12

No team meeting, weekly review, retro or planning cadence is written in any source. memory brew-handbook

Tomorrow

Item Version Tag
Continuous deploy, QV rounds, QA sweeps v1 Stays
A weekly review, 30 minutes v1 New, see Open
A monthly ops check: restore test, spend, secrets v1 New
A handbook snapshot after each weekly review v1 New

How we do it

Proposed weekly review, Monday, 30 minutes, Param, Sai and Arvind:

  1. Next: what got done, what is stuck, what is new. 10 minutes.
  2. Design partners: one line each on Arata, Nutrova, Mana, PUMA India, Scrubroom. 10 minutes.
  3. Decisions needed: each item in Open that waits on Param, answered or given a date. 10 minutes.
  4. After: Claude writes the decisions into the decisions log and takes a handbook snapshot.

Proposed monthly ops check, first Monday:

  1. Restore test, see Restore.
  2. Last month's spend, see Costs and quotas.
  3. Secrets owners and rotation dates, see Secrets and keys.

Numbers

Metric Definition Where measured
Weekly reviews held Reviews that happened, out of weeks Snapshots, Proposed
Decisions per week Rows added to the decisions log Decisions log, Proposed
QA rows open Unfixed rows in Sai's QA sheet The sheet, Proposed

Open

  1. Hold a weekly review? Proposal: yes, Mondays at 11 am IST, 30 minutes, agenda as above, starting 2 Nov 2026.
  2. How often do design partner audits run? Proposal: one full audit across the five partners every two weeks, run by Claude, triaged by Param and Sai.
Sources
  • brew origin/main a559de6: CLAUDE.md 'Release rhythm', digest.js:1-12
  • brew/.wt-prod/docs/production.md:43
  • ~/.claude/scheduled-tasks/handbook-morning/SKILL.md
  • memory brew-qv-process, brew-qa-sweep-2026-10-19, brew-qa-audit-2026-10-20 (index line), brew-qa-round2-2026-10-24 (index line), always-deploy-to-test, brew-prod-backend, brew-handbook

Source file content/people-and-ways-of-working/rituals.md · For agents: Markdown · JSON