[mayteio]

Problem

• Transactional email API callers wait ~900ms (p50) for {queued: true}, ~800ms of it after the job is queued.
• cdp-api awaits two WarpStream acks for telemetry first; a slow Kafka response slows every send.

Changes

• The 202 returns after the job insert and idempotency mark. Telemetry rows go to the producer, which flushes on shutdown; acks finish in the background.
• The email worker counts triggered once, from a triggerMetricPending marker the API sets. Unmarked jobs from old pods aren't counted, so a rollout can dip the count but never double it.

flowchart LR
  subgraph Before
    A1[Insert job] --> B1[Await results + status acks] --> C1[Await app metric ack] --> D1[202]
  end
  subgraph After
    A2[Insert job] --> D2[202]
    A2 -.-> E2[Rows to producer, acks in background]
  end
  classDef phYellow fill:#f9bd2b,stroke:#f9bd2b,color:#000;
  classDef phRed fill:#f54e00,stroke:#f54e00,color:#fff;
  classDef phGray fill:#e5e7eb,stroke:#c7ccd1,color:#000;
  class D1,D2 phYellow;
  class B1,C1 phRed;
  class A1,A2,E2 phGray;

Loading

How did you test this code?

Test rationale: a hanging telemetry flush still gets a 202 (times out with the old await). A marked job counts once across a reschedule; unmarked jobs don't count.

Not checked: latency in prod.

Release status

• No feature flag controls this change

Docs update

• None.

🤖 Agent context

Autonomy: Human-driven (agent-assisted)

Agent: Claude Code, Claude Opus 5.5 (claude-opus-5-5)

Skills: /writing-tests, /writing-code-comments, /writing-pr-descriptions

More context for AI agents

• Decisions: the route awaits handOffQueuedRows() (gzip only, no network), not the acks.
• Follow-ups: an alert on messaging_transactional_email_stage_duration_seconds{stage="total"} p95 once this ships.
• Read closely: cdp-api.ts around the metrics_flush stage, and countPendingTrigger in the email worker.

---

Created with PostHog Desktop

🤖 Generated with Claude Code

[trunk-io]

😎 Merged successfully - details.

[github-actions]

🤖 CI report

✅ Trunk lane — non-backend lane

This PR is assigned to the non-backend lane. It does not run backend Python tests and may merge in parallel with PRs in other lanes.

✅ Duplication (Python) — clean

New Python code duplication introduced by this branch. Fails at 70+ tokens in app code, or 150+ tokens when both copies live in test files. Advisory while the gate proves itself: extract a shared helper instead of copying.

✅ Duplication (TypeScript) — clean

New TypeScript code duplication introduced by this branch. Fails at 70+ tokens in app code, or 150+ tokens when both copies live in test files. Advisory while the gate proves itself: extract a shared helper instead of copying.

🚨 Comment density — 9% of added code lines are comments (16 of 175)

This section warns when comments are more than 3% of the code lines a PR adds, and alerts above 6%. Before agent-assisted PRs, the typical share was about 2%. Only full-line comments count. Docstrings, generated files, snapshots, migrations, and workflow files are left out.

Comments that restate the code, record how the change came about, or narrate the next line add noise for the next reader. Keep the comments that explain a reason the code cannot show, and remove the rest. See .agents/skills/writing-code-comments/SKILL.md for the house rules.

Files with the most added comment lines:

File	Comment lines	Added lines
nodejs/src/cdp/services/monitoring/hog-invocation-results.service.ts	5	37
nodejs/src/cdp/cdp-api.ts	3	12
nodejs/src/cdp/consumers/cdp-cyclotron-worker-email.consumer.ts	3	25
nodejs/src/cdp/consumers/cdp-cyclotron-worker-email.consumer.test.ts	2	68
nodejs/src/cdp/types.ts	2	3
nodejs/src/cdp/cdp-api.serial.test.ts	1	29

This check does not block merging. It updates on every push and clears when the share drops.

[parameterai]

Risk: No findings

The only change since the last review is a no-op refactor in hog-invocation-results.service.ts: replacing rows.map returning undefined entries with rows.flatMap returning empty arrays inside handOffQueuedRows. Behavior, index alignment, and error handling are unchanged, and no security-relevant surface is touched.

_{Sentinel reviewed 6c1260c · Review settings}

[posthog]

PostHog Review · Standard · reviewed 326775b: no findings.

This Standard review started from the posthog-review label. Comment @posthog review --deep for a Deep review, or @posthog help to see all commands.

[trunk-io]

   

View Full Report ↗︎ ⋅ Docs

A new stamphog review started for this PR — the fresh verdict replaces this approval.

[stamphog]

Approved.

The risky write-path and persisted-state changes have owning-team authorship as assurance, with targeted regression tests and no outstanding substantive concerns. The supplied bot review is on an older commit and does not count as a current-head review.

• Author wrote 69% of the modified lines and has 37 merged PRs in these paths (familiarity MODERATE).

Gate mechanics and policy version

Gate		Result
prerequisites	✓	all clear
deny-list	✓	no deny categories matched
size	✓	118L, 5F substantive, 226L/7F incl. docs/generated/snapshots — within ceiling
tier	✓	T1-agent / T1c-medium (226L, 7F, single-area, fix)
stamphog 2.6.1		.stamphog/policy.yml @ 6c1260c · reviewed head 6c1260c

[hosthog]

HostHog preview — posthog-desktop-web

The previews for this PR have been torn down and no longer serve.

[mayteio]

/depot retry

[mayteio]

/depot retry

[deployment-status-posthog]

Deploy status

Environment	Status	Deployed At	Workflow
dev	✅ Deployed	2026-10-29 17:41 UTC	Run
prod-us	✅ Deployed	2026-10-29 18:04 UTC	Run
prod-eu	✅ Deployed	2026-10-29 18:00 UTC	Run