By PromptLeadz · Reading time 16 minutes · 50 prompts across 5 categories · Calibrated for 2026 frontier models

What separates the EM from the EM-influencer

Engineering management is one of the most LinkedIn-saturated disciplines in tech. Threads about psychological safety, 10x engineers, servant leadership, radical candor, and engineering excellence get tens of thousands of likes. The threads describe a vibe. The actual job is memo work.

An engineering manager's primary job is producing decision-grade artifacts: 1:1 prep that surfaces real issues, performance reviews that name specific behaviors, architecture decision records that capture the tradeoffs, post-mortems that name the cause without blaming the person, calibration memos that defend specific reports against peer-manager bias. None of these artifacts look exciting on a screenshot. All of them compound.

Six dimensions separate the EM voice from the EM-influencer voice. Substance: the EM names the specific person, project, or decision; the influencer names the disposition (high-performing, customer-obsessed, founder-mode). Tradeoffs: the EM names what the team is losing with each decision; the influencer says "and" instead of "or". Numbers: the EM opens with the capacity math, the SLA, the latency target; the influencer opens with the team-vibes story. Ownership: the EM names the owner and the date; the influencer names the team. Tone: the EM writes flat memos that hold up under peer review; the influencer writes narrative arcs that go viral. Audience: the EM writes for the team, the calibration committee, and the post-mortem reader; the influencer writes for the algorithm.

Both voices exist in the wild. Only one survives the calibration committee, the post-mortem review, and the third quarterly business review. This pack is calibrated for the first; it explicitly rejects the second at the prompt level by banning the genre's signature phrases inline. Output reads like a memo from an EM who has run a calibration cycle, not a thread from a personal-brand EM who has not.

Five categories. The EM workflow end to end.

The five categories map to the five operating disciplines that determine whether an engineering team compounds or accumulates dysfunction. One-on-Ones and Performance Management comes first because the people work is what compounds; bad 1:1s and vague performance reviews produce the dysfunction other disciplines cannot fix. Technical Leadership and Architecture comes second because the technical decisions an EM signs off on (debt, ADRs, build-vs-buy, on-call design) determine velocity and reliability. Team Health and Process comes third because the team-level discipline (retros that produce change, meeting hygiene, hiring loop quality) is what scales the team beyond founder-EM heroics. Cross-functional and Communication comes fourth because the disagreement with PM, the post-mortem, the all-hands message, the difficult message to upper management are where EMs earn or lose trust. Career and Hiring comes fifth because hiring is where bias and inconsistency creep in, and personal career management is the work most EMs neglect.

Most EMs who fail to compound do so by skipping the unglamorous categories: PIP discipline, capacity math, post-mortem honesty, calibration prep, hiring rubrics. The thread-genre EM skips these in favor of "engineering culture" content; the actual EM does these because they are the leverage.

1. Weekly 1:1 prep memo

Direct report: [paste name, role, level, tenure]. Last 1:1 topics: [paste]. Open threads: [paste]. Recent work: [paste]. Draft a 250-word 1:1 prep: the two questions for them (their priorities, blockers), the two updates from me (calibration, decisions made), the one growth conversation (specific feedback with a recent example), the action items from last week confirmed. 1:1s that drift across status updates without a growth conversation waste both people's time.

2. Performance review that names the actual issue

Person: [paste name, role, level, tenure]. Performance reality (private): [paste honest read]. Goals last cycle: [paste]. Draft a 500-word review: the specific behaviors meeting / not meeting the bar (with examples not adjectives), the bar itself defined, the trajectory (improving, flat, declining), the specific commitments for next cycle, the implicit decision (continue, formal plan, exit). Reviews that praise generally and criticize vaguely produce no behavior change.

3. Promotion case memo

Person: [paste name, current level, target level]. Tenure at level: [paste]. Evidence of next-level work: [paste]. Gaps: [paste]. Draft a 600-word promotion memo for the calibration committee: the named projects demonstrating next-level scope, the named decisions demonstrating next-level judgment, the gaps with the closing plan, the comparison to recent promotions at the same level, the recommendation. Promotion memos without explicit next-level evidence produce calibration friction.

4. Performance improvement plan (PIP) drafting

Person: [paste]. Issues: [paste with specific examples and dates]. Tenure: [paste]. Prior conversations: [paste history]. Draft a 600-word PIP draft: the gap from the bar named with specific examples, the measurable improvements expected (tied to behaviors not outcomes), the support I will provide (coaching, training, shadowing), the timeline (typically 30/60/90), the decision point and the alternative outcome. PIPs written defensively produce legal documents not improvement; this version names the change required.

5. Hard conversation prep with engineer

Person: [paste]. Issue: [paste, e.g. quality, attitude, collaboration, technical judgment]. Context: [paste, what has been tried, who else is affected]. Draft a 400-word prep: the opening that names the issue without rehearsing it, the three to four specific examples ready, the boundary (what is not changing), the path forward, the response if they disagree, the response if they ask 'am I being fired'. Hard conversations rehearsed as scripts often fail.

6. Skip-level meeting prep

Skip: [paste their team, level, tenure]. Stated topic: [paste]. Likely real topic: [paste]. Draft a 250-word skip-level prep: the open question to start (broad, not 'how is it going'), the listen-for items (manager-related signals, team friction, calibration concerns), the things I will NOT promise (changes outside my scope), the action coming out of the meeting. Skip-levels that produce listening without follow-through erode trust faster than skipping them.

7. Backfill vs role redesign decision

Departing role: [paste]. Reason for departure: [paste]. Team gaps: [paste]. Open headcount: [paste]. Draft a 400-word memo: the question of whether the role should be backfilled at all (would the work survive consolidation), the level question (same, more senior, more junior, IC vs manager track), the team-shape implications, the budget implication, the recommendation. Default-backfill produces org bloat; this prompt forces the question of whether the role still needs to exist.

8. Career growth plan with engineer

Person: [paste, current level, time at level, growth signals]. Their stated goals: [paste]. Honest read on trajectory: [paste]. Draft a 500-word growth plan: the next-12-months target (specific level or scope, not 'grow'), the projects that would build evidence, the manager support I will provide, the engineer ownership required, the calibration check-ins. Growth plans without specific projects and check-ins become wishlists.

9. Comp band review for direct report

Person: [paste, level, tenure, performance]. Current comp: [paste]. Market data: [paste source]. Internal compression check: [paste team comps]. Draft a 300-word comp memo: the gap from band midpoint, the recommendation (hold, increase X percent, restructure), the timing (next cycle vs now), the rationale tied to performance and market not tenure, the manager-of-manager approval needed. Comp band drift produces retention surprises at the worst time.

10. Manager calibration prep with peers

Calibration cycle: [paste]. Reports: [paste with current rating]. Outliers: [paste who I want to push for higher or lower]. Draft a 400-word calibration prep: the cases I will defend (with the named evidence), the cases I will accept calibration on, the questions I have about peer reports, the team-level pattern I want to surface. Calibration without prep produces manager-by-manager bias rather than team-level fairness.

11. Technical debt prioritization memo

Current technical debt items: [paste with size estimate and impact]. Engineering capacity: [paste]. Draft a 500-word debt memo: the top 3 debt items by ROI (cost to fix vs cost of carrying), the items to defer with named conditions, the relationship between debt and current product roadmap, the percent of capacity to allocate to debt next quarter, the metric to track. Tech debt prioritized by 'engineering wants' produces low-ROI cleanup; this version forces business ROI.

12. Architecture decision review (ADR)

Decision: [paste]. Context: [paste, why now, who is asking]. Options: [paste top 3]. Draft a 600-word ADR: the question being decided in one sentence, the options with named tradeoffs (performance, complexity, cost, reversibility), the recommended option, the assumptions that, if wrong, would change the decision, the reversibility analysis, the review date. ADRs without named reversibility produce decisions that get re-litigated when assumptions shift.

13. Build vs buy decision for engineering tooling

Need: [paste, e.g. observability, deployment, IDE, internal tool]. Build estimate: [paste people-weeks and ongoing]. Buy options: [paste vendor, pricing, fit]. Draft a 500-word memo: the explicit cost comparison (build first-year, build ongoing, buy first-year, buy ongoing), the team capacity tradeoff (engineers building tooling vs shipping product), the strategic question (is this our differentiator), the recommendation, the kill criterion if buy fails. Build-vs-buy in engineering typically defaults to build (NIH bias); this prompt forces the question.

14. On-call rotation design

Team size: [paste]. Current rotation: [paste]. Pain points: [paste]. Service criticality tier: [paste]. Draft a 500-word on-call memo: the rotation structure (primary, secondary, escalation, business hours vs 24/7), the load per engineer per month (target: humane), the comp adjustment (additional pay or time off), the runbook discipline required, the handoff process. On-call rotations that overload senior engineers produce attrition.

15. Code review standards for the team

Team: [paste]. Review pain points: [paste, e.g. slow turnaround, inconsistent quality, nit-fights]. Draft a 400-word standards memo: the scope (what reviews focus on: correctness, security, performance, design, not style), the SLA for review turnaround (specific business hours), the 'approve with comments' threshold, the disagreement resolution (when authors and reviewers disagree, who decides), the auto-formatter and lint discipline. Review standards that try to cover everything produce nit fights; this version names what reviews are for.

16. Incident response runbook

Service: [paste, criticality tier]. Recent incidents: [paste]. Current runbook state: [paste]. Draft a 600-word runbook: the detection paths (alerts, customer reports, monitoring), the first responder actions in 5 numbered steps, the escalation criteria with named owners, the comms responsibilities (internal vs external, who writes the status page), the post-incident review trigger, the runbook review cadence. Runbooks written during incidents fail.

17. Migration plan memo

Migration: [paste source, target, scope]. Current state: [paste]. Risks: [paste]. Draft a 700-word migration memo: the phased plan (typically: read, dual-write, switch reads, switch writes, retire), the rollback criteria at each phase, the customer or service impact, the timeline with specific milestones, the abandonment criteria (when to stop and roll back permanently). Migration memos without explicit rollback criteria produce stuck migrations.

18. Performance investigation framework

Performance issue: [paste, e.g. p99 latency, throughput, memory]. Service: [paste]. Recent changes: [paste]. Draft a 400-word framework: the hypothesis prioritization (recent change, dependency, traffic shift, infrastructure), the measurement plan per hypothesis, the time-boxed investigation (typically 2-3 days before escalating), the rollback criteria, the longer-term mitigation. Performance investigations that skip the hypothesis stage produce chase-the-symptom debugging.

19. Vendor security review for engineering tools

Vendor: [paste, use case]. Data accessed: [paste]. Existing security review: [paste status]. Draft a 400-word security memo: the data sensitivity classification, the security review status (SOC 2, ISO, custom), the access controls (SSO, scoped tokens, audit logs), the contract terms required, the security team approval needed. Security reviews skipped at procurement produce incidents at the worst time.

20. Engineering capacity planning memo

Team headcount: [paste, with roles and experience]. Quarter ahead: [paste roadmap items with estimates]. Operational load: [paste, e.g. on-call, support, maintenance]. Draft a 500-word memo: the realistic capacity (typically 60-70 percent of headcount-weeks, not 100), the roadmap that fits the capacity, the items that don't fit (and the explicit prioritization), the operational load named separately, the contingency for unknowns. Capacity plans that assume 100 percent project work miss reality.

Most EM advice circulating on LinkedIn is content marketing. The work that actually compounds is the work that does not screenshot well.PromptLeadz Engineering Manager Pack

21. Quarterly retro that produces change

Quarter: [paste]. Major events: [paste, e.g. launches, incidents, hires, losses]. What worked: [paste]. What did not: [paste]. Draft a 500-word retro: three things to keep doing (named at the action level, not 'good comms'), three things to change (with the specific change, owner, timeline), one thing to stop entirely, the metric that would tell us the change worked, the next retro date. Retros that produce 'communicate better' as outcome produce no change.

22. Meeting hygiene policy for the team

Current meeting load: [paste, e.g. average meetings per week, recurring]. Friction: [paste]. Draft a 500-word policy memo: the specific policies (agenda required 24 hours before, decision-or-cancel rule, default 25/50 minutes not 30/60, recurring meetings auto-cancel quarterly unless re-justified), the enforcement mechanism, the meeting types this excludes (1:1, all-hands), the rollout sequence. Meeting policies unenforced produce no change.

23. Team morale check

Team size: [paste]. Recent events: [paste, e.g. launch, layoff, leadership change, incident]. Observable signals: [paste]. Draft a 400-word morale assessment: the leading indicators (1:1 tone shifts, code review snark, late-night Slack patterns), the lagging indicators (attrition risk, engagement scores), the specific actions in the next 30 days (not 'improve culture'), the escalation if signals worsen. Morale assessed once a year through engagement surveys misses the early signals.

24. Hiring loop design

Role: [paste]. Level: [paste]. Existing loop: [paste]. Pain points: [paste]. Draft a 500-word loop design: the screening criteria (specific, measurable), the on-site stages (typically: technical interview 1 and 2, design, manager, bar-raiser), the rubric per stage, the debrief process, the bias mitigation. Hiring loops without explicit rubrics produce vibes-based decisions and disparate outcomes.

25. Bar raiser and debrief discipline

Recent hire decisions: [paste with outcomes]. Disagreement patterns: [paste]. Draft a 400-word debrief memo: the structure (each interviewer presents independently before the discussion, no peeking at scores), the bar raiser role and authority, the disagreement resolution (typically: bar raiser veto plus manager strong signal), the data we capture for hiring quality measurement. Debriefs that drift to consensus before independent presentation produce groupthink hiring.

26. New hire onboarding 30/60/90

New hire: [paste name, role, level]. Manager: [paste]. Team context: [paste]. Draft a 500-word onboarding plan: the 30-day milestones (relationships, environment, first PR), the 60-day milestones (first project owned), the 90-day milestones (first measurable contribution), the support cadence (weekly 1:1, manager check-ins, peer buddy), the success criteria. Onboarding plans without specific 90-day output produce drift in the first quarter.

27. Engineering team offsite design

Team: [paste size, distributed status]. Goals: [paste]. Budget: [paste]. Draft a 400-word offsite plan: the work-vs-bonding split (typically 70/30 for a quarterly), the structured sessions (architecture review, retro, planning), the unstructured time (deliberate, not vague), the location and accommodation logic, the success measure 90 days later. Offsites without measurable goals produce expensive bonding events that don't move the team forward.

28. Slack hygiene for engineering team

Channel structure: [paste]. Pain points: [paste, e.g. notification fatigue, decision drift, archive sprawl]. Draft a 300-word Slack memo: the channel-purpose discipline (each channel has one stated purpose), the threading rules (replies in threads, not main channel), the at-channel and at-here policy (rare, with clear criteria), the archive cadence. Slack hygiene unenforced produces 50-channel sprawl and notification fatigue.

29. Documentation expectations for the team

Documentation surface: [paste]. State: [paste honest assessment]. Draft a 400-word memo: the expectations per artifact type (design docs, runbooks, READMEs, ADRs), the maintenance ownership (every doc has one owner), the review cadence (typically 6-month freshness audit), the deprecation process (delete vs archive). Documentation expectations without ownership produce stale wikis.

30. Time-zone and async work norms

Team distribution: [paste time zones]. Current async vs sync split: [paste]. Pain points: [paste]. Draft a 400-word norms memo: the meetings that must be sync (rare, named), the work that must be async (most), the response-time SLA per channel (Slack vs email vs PR review), the overlap window for cross-zone collaboration, the doc-driven decision discipline. Async work without explicit norms produces sync defaults and burnout.

31. Disagreement with PM on roadmap

Disagreement: [paste, e.g. priority, scope, timeline]. PM position: [paste]. My position: [paste]. Stakes: [paste]. Draft a 500-word resolution memo: the actual disagreement named (often not the surface issue), the underlying tradeoff (resource, technical risk, customer signal), the decision-maker (engineering, product, both via head of), the proposal with rationale, the implementation path, the review date. PM-EM disagreements resolved by the loudest voice produce repeated cycles.

32. Stakeholder update for cross-functional initiative

Initiative: [paste]. Stakeholders: [paste roles]. Status: [paste honest read]. Draft a 500-word update: the headline (one sentence), the milestones hit (with dates), the milestones missed (with cause and revised date), the risks emerging, the asks of stakeholders, the next checkpoint. Stakeholder updates that perform progress without naming risks lose trust over multiple cycles.

33. Pre-mortem for engineering project

Project: [paste]. Timeline: [paste]. Stakes: [paste]. Draft a 400-word pre-mortem: 'imagine this project failed in 6 months: what would have caused it', the top 5 risks ranked by likelihood and impact, the mitigation per risk, the kill criterion (the metric or signal that would tell us to stop or pivot), the decision-maker for the kill. Pre-mortems run before launch surface risks the team is reluctant to name during launch enthusiasm.

34. Post-mortem that names the cause not the person

Incident: [paste, e.g. outage, customer escalation, missed deadline]. Impact: [paste]. Draft a 600-word post-mortem: the timeline of events with timestamps, the root cause (technical, process, decision, judgment, named without blame), the contributing factors, the decisions that would have prevented it (with the trade-off named), the action items with owners and dates, the system change to prevent recurrence. Post-mortems that name a person produce defensive culture; that blame 'process' produce no change.

35. Difficult message to upper management

Topic: [paste, e.g. miss, escalation, key person event, pushback on direction]. Audience: [paste, e.g. director, VP, CTO]. Draft a 400-word message: the news in the first sentence (no preamble), the cause (factual, not narrative), the response (specific actions taken), the implication for plans, the ask if any. Messages to upper management that bury news in narrative produce them hearing it secondhand.

36. Customer-facing engineering communication

Topic: [paste, e.g. outage post, deprecation, breaking change]. Audience: [paste customer segment]. Draft a 400-word communication: the news named clearly (first paragraph), the impact (specific not vague), the timeline for resolution or migration, the support resources, the contact for follow-up. Customer-facing engineering comms that obscure the impact produce trust loss.

37. Engineering vs sales conflict mediation

Conflict: [paste, e.g. sales committed to feature engineering can't ship, post-sale promises]. Sales position: [paste]. Engineering position: [paste]. Draft a 400-word mediation memo: the surface conflict and the underlying root cause (typically GTM and engineering planning misalignment), the decision-maker (who has authority to commit features), the immediate resolution for the named case, the systematic change to prevent recurrence. Sales-engineering conflicts resolved one deal at a time produce repeated cycles.

38. Engineering input on product strategy

Strategy under discussion: [paste]. Engineering view: [paste]. Draft a 500-word engineering perspective memo: the technical feasibility honestly assessed (not optimistic, not pessimistic), the resource implications (team capacity, hiring, time-to-value), the technical risks the strategy creates, the technical opportunities it enables, the recommendation for sequencing. Engineering input that just says 'feasible with effort' adds no signal; this version names specific risks and opportunities.

39. Engineering all-hands message

Topic: [paste]. Audience: [paste team size]. Draft a 400-word all-hands message: the news in the first sentence, the context (what changed, what didn't), the implication for the team (work, structure, expectations), the ask (what the team should do, where to ask questions), the timeline for next update. All-hands messages that bury news in narrative produce rumor.

40. Cross-team RFC review

RFC: [paste, scope, author team]. Stakeholders: [paste]. Draft a 400-word review: the scope and the named decision being requested, the technical evaluation (correctness, design, alternatives considered), the cross-team implications (downstream effects, ownership questions), the open questions, the recommendation (approve, approve with conditions, request revision). RFC reviews focused only on technical correctness miss cross-team implications.

41. Self-assessment for own performance review

Cycle: [paste]. My role and tenure: [paste]. Goals from last cycle: [paste]. Outcomes: [paste]. Draft a 500-word self-assessment: the goals hit (with specific evidence), the goals missed (with cause and what I learned), the projects I led (with team-level outcomes), the gaps I am working on, the next-cycle priorities. Self-assessments that perform humility without specifics fail to make the case for the next role.

42. Asking for the promotion

Current level: [paste]. Target: [paste]. Tenure at level: [paste]. Evidence I have gathered: [paste]. Draft a 500-word memo to my manager: the concrete request (specific level, specific timeline), the named projects demonstrating next-level scope, the named decisions demonstrating next-level judgment, the gaps and how I am addressing them, the ask (their support, their advocacy, calibration). Promotion asks that come as 'do you think I am ready' produce vague answers.

43. Career conversation with my own manager

My role: [paste]. Time at level: [paste]. Aspirations: [paste]. Draft a 400-word conversation prep: the specific career question (path, pace, scope), the evidence I will bring, the listen-for items (their honest read on my readiness, the specific gaps, the org constraints I should know), the action coming out. Career conversations that drift through vibes produce vibes-based career outcomes.

44. Considering a role change (internal)

Current role: [paste]. Target role: [paste]. Reason: [paste, e.g. growth ceiling, scope mismatch, manager change]. Draft a 400-word memo for my own clarity: the reasons honestly named, the network reads I need (skip-level, target manager, peer in target role), the timing implications (cycle, deliverables in flight), the relationship-with-current-manager management. Internal role changes done badly burn relationships.

45. Considering leaving the company

Tenure: [paste]. Trigger: [paste, e.g. burnout, comp, scope, manager]. Draft a 500-word personal memo (not for sharing): the trigger named honestly, the gap from what would keep me, the realistic alternatives (similar role elsewhere, different role, sabbatical), the financial and career calculus, the conversations to have before deciding. Quitting decisions made in 24 hours produce regret.

46. Hiring panel rubric design

Role: [paste]. Level: [paste]. Skills required: [paste]. Draft a 600-word rubric: the dimensions evaluated (typically: technical depth, problem-solving, communication, collaboration, judgment), the leveling guide per dimension (what does meets-bar look like), the calibration examples (specific signals that fit each level), the disqualifying signals (rare, named), the scoring scale. Rubrics without specific leveling guides produce score bias and consistency drift.

47. Reference check for senior hire

Candidate: [paste, role under consideration]. Reference: [paste]. Goal: [paste, e.g. validate technical depth, validate management ability, validate ethics]. Draft a 300-word prep: the framing of the call, the 5 specific questions per goal, the listen-for items (hesitations, off-record asides), the closing that earns honesty (e.g. 'would you hire them again given the chance'). Reference calls asking 'what is it like working with them' produce no signal.

48. Interview kit for technical interview

Role: [paste, level]. Topic: [paste, e.g. system design, coding, architecture review]. Draft a 600-word interview kit: the prompt with progressive depth (start broad, drill down), the rubric mapped to leveling, the follow-up probes, the calibration of 'meets bar' vs 'above bar', the time allocation. Interview kits without leveling rubrics produce interviewer-dependent scoring.

49. Internal mobility decision

Person: [paste, current role, requesting move to: paste]. Manager-in-target-role read: [paste]. Performance current: [paste]. Draft a 400-word memo: the case for the move (skill match, growth, retention), the case against (current team gap, performance, fit), the trade-off named, the recommendation, the transition plan. Internal mobility decisions that prioritize retention over fit produce damaged target teams.

50. EM-to-EM peer conversation prep

Peer: [paste team]. Topic: [paste, e.g. cross-team coordination, calibration alignment, hiring philosophy]. Draft a 300-word prep: the specific question I am bringing, the data I have on it, the listen-for items from their team, the action coming out. EM peer conversations that drift across status updates miss the structural-improvement opportunity.

How the prompts fit a real EM week and quarter

Daily: 1:1 prep notes, hard-conversation prep when needed, code review and ADR review. The daily memo discipline is what compounds.

Weekly: capacity check against the week's commitments, stakeholder update for in-flight initiatives, retro action item check. The weekly cadence catches drift before it becomes a quarterly miss.

Monthly: team morale check, comp band review, hiring loop debrief discipline, cross-team RFC review. The monthly cadence is where team health gets diagnosed.

Quarterly: performance reviews, calibration prep, retro that produces change, capacity planning, technical debt prioritization. The quarterly cadence is where the EM job actually shows.

Annually: career conversation with own manager, self-assessment, hiring panel rubric review, on-call rotation rebalance. The annual cadence is where the EM career gets managed deliberately.

A good EM 1:1 produces a memo. A good EM performance review produces a memo. A good EM ADR produces a memo. The job is memo work. The threads about the job are not.PromptLeadz Engineering Manager Pack, Section 6

Five mistakes that wreck EM prompts

1. Filling in the prompt with vibes instead of specific examples and dates. The prompts ask for named behaviors, specific dates, exact levels. Filling with 'strong', 'high-performing', 'needs work' produces output of the same low calibration. The discipline is putting the actual behaviors and dates into the inputs.

2. Treating the output as the final memo. The prompts produce drafts. The actual memo is the draft after you have edited it for accuracy, removed the LLM-cliche phrasing the humanizer step would catch, and verified that every example matches reality. Shipping the LLM draft directly is sloppy.

3. Skipping the prompts that ask uncomfortable questions. The PIP draft, the considering-leaving memo, the cap table for an underperformer's exit. The avoided prompts are usually the ones with the most leverage. Notice the avoidance.

4. Sharing the LLM draft externally without redaction. The prompts produce internal artifacts naming specific people, projects, and decisions. The outputs should not leave the management chain without explicit review.

5. Running the EM-influencer prompts instead of these. Prompts that produce 'engineering excellence' content reinforce the genre this pack rejects. Calibration to the LinkedIn-thread voice produces threads, not memos.

Sources and further reading

The pack draws on a body of public work from senior engineering leaders. Recommended reading for EMs who want depth beyond the threads.

Will Larson's writing at lethain.com is the most rigorous public body of engineering management writing. The book An Elegant Puzzle is the foundational text. The book Staff Engineer covers the IC-track parallel.

Camille Fournier's The Manager's Path is the single best book for new and progressing engineering managers, especially the chapters on tech lead, manager, manager of managers, and director.

Gergely Orosz's The Pragmatic Engineer at pragmaticengineer.com covers the realities of engineering at scale, including compensation, hiring, and team structure data.

Lara Hogan's writing on management, feedback, and difficult conversations is the most practical public work on the people side of engineering management.

About PromptLeadz

PromptLeadz publishes free component-built prompt packs and the production-grade Drop-in utilities that wrap them. The franchise covers role-based packs (PM, EM, CSM, Sales Leader, Operator, Data Analyst, VC), format-based packs (.md agent files in breadth and depth), and the underlying frameworks (the 8-Component Skeleton, the Anti-Prompt-Engineering Manifesto).

Every pack rejects the LinkedIn-influencer voice at the prompt level by banning the genre's signature phrases inline. The result is output calibrated for memos that survive peer review, not threads that go viral. Free packs ship with no email gate at promptleadz.com.

Questions people ask

Who is this engineering manager prompt pack for?

Engineering managers, senior engineering managers, directors of engineering, and tech leads acting in a manager capacity. Most useful for EMs of teams 4 to 30, in product engineering, infrastructure, platform, or applied ML. The prompts assume basic EM literacy: 1:1 cadence, performance review process, ADR discipline, on-call rotations, and hiring loops.

Does it work for EMs at large companies versus startups?

Yes for both, with calibration. Large company EMs lean on the calibration, promotion case, and PIP prompts. Startup EMs lean on the architecture decision, on-call design, and capacity planning prompts. The performance review and 1:1 prompts work in both environments. The hiring loop prompts assume some structure exists.

Why does the pack ban phrases like 10x engineer and psychological safety?

Both phrases are legitimate concepts that have been ground into LinkedIn cliches. The pack bans the cliche framing because it produces low-calibration output that performs care or excellence rather than naming the actual issue. Real EMs talk about specific behaviors and specific tradeoffs; the cliche genre talks about culture and excellence in general.

What output format do the prompts produce?

Memo register: flat, factual, named tradeoffs, specific owners and dates. The opposite of LinkedIn-thread register. The prompts are calibrated for internal use: 1:1 prep notes, performance review drafts, ADRs, post-mortems, and calibration memos that hold up under peer review.

How does this pair with other PromptLeadz packs?

Pairs with the 8-Component Skeleton framework as the foundation, the Operator Pack for the cross-functional and capacity planning prompts, and the Founder Pack for early-stage EMs reporting directly to founders. The Anti-Prompt-Engineering Manifesto explains why component-built prompts beat magic-words prompts for serious work.

Are these prompts safe to share with my team?

The prompts themselves are free to share. The outputs of the performance, calibration, PIP, and hiring prompts are confidential and should not leave the management chain. The prompts assume the discipline of internal-only handling for sensitive output.

Do these prompts work with Claude, ChatGPT, and Gemini?

Yes for all three. The prompts are built on the 8-component skeleton (identity, context, task, constraints, examples, output format, refusal conditions, evaluation) which works across frontier models. Claude tends to handle the memo register most naturally; ChatGPT requires slightly tighter constraints; Gemini works best when the output format is named explicitly.

What is the difference between an EM and a senior engineer in this context?

An engineering manager is responsible for the people, the process, and the technical outcomes of a team, while a senior engineer is responsible for technical outcomes and technical leadership without direct people management. The pack is calibrated for managers; senior engineers will find the architecture, code review, and on-call prompts useful but should pair with an IC-track prompt set for career and technical depth work.

The franchise: free packs, frameworks, and the manifesto

The thesis behind all of them: The Anti-Prompt-Engineering Manifesto. The framework: The 8-Component Skeleton.

Free packs, no email gate · Calibrated for 2026 frontier models · promptleadz.com