Feedback that lands after incidents

If Feedback that lands after incidents only happens in 1:1s that never change assignments, you are performing care. Real mentorship changes what someone owns next week.

Growing seniors who can grow others is how organizations scale craft without scaling heroics.

I will stay on this thesis — no adjacent manifesto, no tour of every neighboring discipline.

Specific feedback tied to a PR, RFC, or incident role compounds. Vague encouragement at review time feels kind and teaches nothing.

Write one page while the decision is still warm — context, options, choice, owner, revisit date. Verbal alignment on Feedback that lands after incidents evaporates under ordinary calendar pressure, and Slack archaeology is a poor substitute for a decision record.

Practices with enough detail to copy

Cadence beats intensity. These hold when they fit inside weeks people already live.

1

When AI drafts: require a human verification note in the PR describing what was checked and how.

Keep a light written trail of growth so progress is not a vibe only the manager remembers. Promotion packets should not require archaeology.

2

Every month: one public sponsorship act — authorship, presentation, or incident lead.

Use review comments as teaching artifacts. Prefer “here is the pattern and why it matters in production” over drive-by nits. A review that only polices style teaches compliance, not craft.

3

Every 1:1: one concrete coaching ask, one ownership shift, one written follow-up.

The psychology of asking for help is shaped by what got rewarded last time someone asked. If asking is punished as weakness, AI tools will hide mistakes instead of surfacing uncertainty.

4

Every PR review: at least one comment that teaches a reusable pattern.

Promotion packets should be assembled from artifacts already in the open: RFCs, incident leads, teaching reviews. If the packet requires secret knowledge, mentorship failed earlier.

Stretch ownership without a safety net is abandonment. Stretch ownership with a named coach, a kill switch, and a debrief is how judgment grows.

Seniors need deliberate practice teaching: lead a review huddle, write the onboarding doc, own feedback quality. If you only reward personal velocity, teaching becomes unpaid overtime.

On Feedback that lands after incidents, care that never changes next week’s ownership is performance. Stretch a thin slice, coach the decision, and leave an artifact someone else can reuse.

Scenes

AI-assisted workflows arrive without coaching norms. People paste secrets into tools, accept insecure snippets, and skip tests because the model “looked sure.” Literacy clinics on one failure mode per month beat a single all-hands about AI.

A junior with a strong copilot ships fluent PRs. Mentorship is teaching them to distrust fluent wrongness: invariants, authz edges, and “what breaks at 10x load.” Without that coaching, you accelerate confident mistakes.

Feedback arrives only at performance review. By then it feels political. The fix is weekly specific notes tied to artifacts — PR comments, decision records, incident roles — so the review is a summary, not a surprise.

Tradeoffs

With AI drafting, the tradeoff is review load. Explicit verification notes reduce merge regret more than asking people to “be careful.”

The tradeoff in not stealing the work is short-term speed. You will be slower this afternoon so the team is faster next month. Leaders who cannot tolerate that tradeoff become permanent critical path.

The tradeoff in public sponsorship is risk to your own reputation. Take it. That is part of the job. Private cheerleading costs you nothing and buys them little.

Seniors need deliberate practice teaching. If the scoreboard only rewards personal velocity, judgment does not scale past one hero.

Measure what you claim to care about. If Feedback that lands after incidents only appears in kickoff slides and never in review, incident, or planning artifacts, it is branding. Put a verification signal where people already look.

Failure modes

Each of these is a missing control, not a personality problem:

• Stealing the work under the banner of helping — the mentee never practices judgment under load.
• AI tools introduced without coaching on verification, data boundaries, or accountability.
• Mentorship as calendar filler: 1:1s that never change assignments or ownership.
• Seniors rewarded only for personal output, so teaching becomes unpaid overtime.

Pair on the decision, not only the code. Have the mentee write the tradeoff paragraph first, then implement. Judgment is a writing skill under time pressure as much as it is a coding skill.

Critique privately and specifically. Vague encouragement teaches nothing. Vague criticism teaches fear. Specific critique tied to a reusable pattern teaches judgment.

Sponsor publicly. Put someone’s name on the design review invite, the RFC author line, the customer-facing postmortem. Private belief in someone does not compound; public sponsorship does.

Feedback that lands after incidents should arrive within days, tied to a specific decision or omission, and paired with a practice rep. Delayed generic feedback teaches nothing useful.

Career conversations should change next week’s work: a stretch ownership, a public design, a safer incident role — not only aspirational titles twelve months out.

Teach the verification step out loud when models draft. Juniors copy what seniors check — or what seniors skip.

When someone freezes asking for help, treat it as a systems smell. Latency to ask questions predicts incident severity. Psychological safety is an engineering control.

Close

Grow judgment in public artifacts. Hallway wisdom does not compound.

Carry the claim as a habit, not a brand: Feedback that lands after incidents either compounds ownership or it was applause.