Content review and open judgments

Reviewed 1 October 2026 after the candidate-facing content critique. All 120 original answer targets, question-specific anchors, and failure modes were examined. This records current editorial judgments, not a reconstruction of unrecorded doubts during the original research or independent rubric validation.

What the critique revealed

The library was organized around topic coverage, with much of the definition of good judgment implicit in experience and scenario answers. Explicit criteria were uneven: M09 asks what makes a team healthy, O03 asks what makes an incident review useful, and W04 asks what makes a decision record useful, but there was no equivalent question about a good decision or a good technical decision. Logistics also passed the selection boundary despite offering little substantive preparation value. Salary advice treated several tactics as equally useful without recommending an information-seeking default.

The repair belongs in question selection and canonical answer guidance. The atlas copies those sources; changing only its appearance would preserve the editorial defect. Counts describe coverage and must never become a target that justifies filler or equal-sized categories.

Changes made

• Retired C08; availability, location, travel, and authorization remain user-owned checklist facts. Its ID is preserved as retired, not recycled.
• Rewrote C07 to seek the employer's budgeted range firmly and persistently before volunteering a candidate number. Pressure, vague bands, package ambiguity, and refusal now have explicit preparation guidance. The final choice remains with the candidate.
• Added B11, “What makes a decision a good decision?”, and TS09, “What makes a technical decision a good technical decision?”, with separate criteria and application probes.
• Added principle/concept answers to evaluation. A definition question needs criteria, limits, and application; it should not be forced into a success story.
• Added synthetic C07 pressure-handling dialogue and B11/TS09 principle answers with counterexamples. There are now 17 families with worked illustrations.

Doubts that remain

These are content review judgments, not claims that the original authors agree.

1. How much of this bank matters for a particular loop? M&A integration, reductions, incident response, platform design, and agent authority are useful in specific roles. They should not make a universal preparation curriculum.
2. Are neighboring cards genuinely distinct? Several have different targets but can use the same episode. Preserve that distinction in evaluation, while avoiding repetitive practice. The walkthrough flags the closest overlaps.
3. Are principles explicit enough? Decision quality is now covered. Good management, staff-level contribution, engineering quality, and organizational effectiveness still deserve an explicit concept-to-example coverage review.
4. Do the answer targets overprescribe process? “Context, options, evidence, owner, checks” is useful but can become a checklist response. A strong answer must choose material criteria and explain a position; every answer need not mention every safeguard or prove that a formal process was used.
5. How much attribution is direct? Sources include direct interview guidance, extrapolated practices, selected passages, and book overviews. Book listings are reading paths, not evidence of complete-book research. Our anchors remain editorial even when the interview theme comes directly from an author.
6. Where do authors disagree? Chemistry versus content, technical involvement, potential versus demonstrated work, and deciding versus delegating have real tensions. Accountable delegation and abdication must not be conflated. The source ledger records the decisions rather than implying consensus.
7. How trustworthy is evaluation? Three dimensions per card are a convenient draft convention, not evidence that every competency has exactly three parts. Only 14 original question families have worked examples. Neither those illustrations nor numeric anchors validate automated grading. Independent calibration, unconventional sound answers, and unsuccessful sound decisions remain necessary before product use.

Question-by-question walkthrough

“Keep” means useful preparation content, not a mandate to practice it. “Select” means choose it for actual responsibilities or a known interview. “Overlap” identifies an adjacent target to keep distinct. A row is an editorial review note; the owning card contains the full answer and rubric.

Common and recruiter

ID	Question	Review judgment
C01	Tell me about yourself	Keep: relevant contribution and direction; avoid resume recitation or a prescribed persona.
C02	Why this role and company?	Keep: informed interest; do not require manufactured passion or insider knowledge.
C03	Why are you leaving?	Keep: concise truthful transition; overlaps C06 but should not demand private disclosure.
C04	What strength would you bring?	Keep: capability plus mechanism; use B01 evidence without repeating the same answer.
C05	What are you improving?	Keep: genuine development with changed behavior; avoid a strategically harmless fake weakness.
C06	Explain this transition or gap	Select when relevant: transferable capability; a gap is not a competency defect.
C07	Compensation expectations	Revised: employer range first, professional persistence, clear package, and truthful response to pressure.
C08	Availability, location, travel, authorization	Retired: checklist facts do not justify a substantive preparation card or scoring dimensions.
C09	What do you want next?	Keep: responsibilities and environment; overlaps career choice LQ08, with a forward-looking target.
C10	Questions for the employer	Keep: due diligence that changes the candidate's decision; not a performance of enthusiasm.

Behavioral

ID	Question	Review judgment
B01	Meaningful accomplishment	Keep: actual contribution and useful effect; achievement alone cannot establish decision quality.
B02	Failure or regretted decision	Keep: responsibility and original information; do not require an eventual triumphant outcome.
B03	Disagreement with a colleague	Keep: represent both positions fairly and reach a decision; distinguish from hierarchy in B10.
B04	Difficult feedback received	Keep: examination and response; accepting every criticism is not the ideal answer.
B05	Working through ambiguity	Keep: diagnosis and bounded progress; avoid rewarding either indefinite research or certainty.
B06	Competing priorities or saying no	Keep: real opportunity cost and communication; overlaps DL02/E03 at different scopes.
B07	Influence without authority	Keep: shared context and adoption; persuasion theater or a circulated document is insufficient.
B08	Changing your mind	Keep: evidence changed a belief; the original choice need not have been wrong.
B09	Crisis or sustained pressure	Select a relevant example; separates general coordination from O01's operational response.
B10	Pushback on a leader	Keep: proportionate dissent and limits; close overlap with K08 should not produce duplicate practice.
B11	What makes a decision good?	Added: explicit criteria, uncertainty, timing, outcomes, and learning; tests the model behind the stories.

Management fundamentals

ID	Question	Review judgment
M01	Why management and your approach?	Keep: role understanding and demonstrated principles; an explicit “good manager” variant is worth reviewing.
M02	How technically involved?	Keep: context-dependent engagement; reject a universal coding percentage.
M03	Delegating important work	Keep: authority, support, and retained responsibility; distinguish delegation from avoidance.
M04	Helping someone grow	Keep: their goals and capability; do not make the manager the author of another person's success.
M05	Underperformance	Keep: expectations, causes, direct dialogue, and accountability; no universal disciplinary script.
M06	Difficult feedback role-play	Keep and exercise interactively later; a generic monologue does not prove responsive judgment.
M07	Useful one-to-ones	Keep, lower priority than actual hard decisions; ritual and fixed cadence are not the outcome.
M08	Building a hiring process	Select for hiring responsibility; distinguish process design from a disputed hiring decision in HR02.
M09	Healthy and effective team	Keep: already an explicit good-looks-like question; balance purpose, capability, candor, and sustainability.
M10	Employee upset about promotion	Keep: individual expectations and fair next actions; HR04 evaluates the broader calibration system.

Staff and principal IC

ID	Question	Review judgment
T01	Significant technical project	Keep: depth, direction, contribution, and adoption; needs a challenging technical follow-up rather than generic polish.
T02	Difficult technical tradeoff	Keep: application in one real choice; TS09 now supplies explicit technical decision criteria.
T03	Debt or large migration	Keep: costs, transition, adoption, retirement; scope is broader than one rewrite story.
T04	Harsh review and junior engineer	Keep: concrete scenario combining technical standards with care; strong direct interview relevance.
T05	Making others effective	Keep: enabling capability and avoiding a permanent bottleneck; do not reward unlimited glue work.

This track is thin relative to management. Role archetypes, senior-versus-staff expectations, choosing leverage, and strategy through influence need a focused follow-up; larger scope or more meetings cannot stand in for staff contribution.

Executive

ID	Question	Review judgment
E01	Develop and evaluate strategy	Keep: diagnosis and consequential choices; explicitly ask what makes a strategy good as a variant.
E02	First 90 days	Keep: targeted learning with proportionate urgency; a rigid calendar is weak preparation.
E03	Resource allocation	Keep: business objective, alternatives, and opportunity cost; must use actual financial authority.
E04	Change the organization	Keep: system diagnosis and design; G06 concerns carrying out the transition.
E05	Engineering value	Keep: a measure informs a decision; no business-causality claim from proximity alone.

Executive breadth is useful but incomplete. CEO/board partnership, functional mandate, financial literacy, and executive-team conflict need role-specific research. They should not be silently represented by generic manager answers.

Delivery and execution

ID	Question	Review judgment
DL01	Estimates and commitments	Keep: basis, confidence, dependencies, and revision; no false precision.
DL02	Fixed deadline, insufficient capacity	Keep: essential outcome and feasible choices; overlaps B06's prioritization at a different target.
DL03	Project recovery	Keep: change a diagnosed cause; changing the date or demanding effort is insufficient.
DL04	Multi-team dependencies	Select cross-team work: feasible agreements and usable ownership, not coordination heroics.
DL05	Process improving delivery	Keep: problem, adoption cost, and observed benefit; no framework compliance scoring.
DL06	Release readiness	Keep: consequences and relevant proof; distinguish from operational capability in O06.
DL07	Stopping after investment	Keep: future value and responsible closure; useful test of sunk-cost reasoning.
DL08	Slow execution versus blocked decisions	Keep: trace actual waiting; do not assume every delay is a decision problem.

Reliability and operations

ID	Question	Review judgment
O01	Organizing incident response	Select operational responsibility: mitigation, roles, and verified recovery.
O02	Reliability versus features	Keep for service leaders: user consequences and explicit tradeoffs; no universal target.
O03	Useful incident review	Keep: already defines good practice; prioritize changed behavior over action-item counts.
O04	On-call overload and toil	Select: diagnosis, relief, and durable shared capability; resilience is not unlimited availability.
O05	Risky rollout and rollback	Select: actual state and recovery mechanisms; “we can roll back” is not enough.
O06	Operational readiness	Select: ownership and proven mechanisms; overlaps DL06 but has a post-release operating target.
O07	Incident communication	Select: verified audience-useful updates; can be a follow-up to O01 when the loop is short.
O08	Security vulnerability	Select: risk and expert partnership; general leadership preparation cannot certify security expertise.

Product and business judgment

ID	Question	Review judgment
P01	Request without a clear problem	Keep: clarify outcome while respecting real urgency and product authority.
P02	Direct customer learning	Select where expected: behavior and sampling limits; engineers need not invent research ownership.
P03	Testing a risky assumption	Keep: material uncertainty and evidence that can change a decision.
P04	Conflicting customer demands	Select commercial exposure: commitments, value, and continuing cost.
P05	Shipped feature unused	Keep: verify measurement and diagnose before adding code; delivery differs from value.
P06	Engineering/product/design decisions	Keep: shared outcomes and accountable closure; avoid one preferred operating-model script.
P07	Technical choice and business viability	Keep at appropriate scope: explain a mechanism and bound causal claims.
P08	Experiment challenges your idea	Keep, or pair with P03: interpretation and stopping; close overlap with B08's belief update.

Managing managers and change

ID	Question	Review judgment
G01	Developing another manager	Select managers of managers: their capability and team effects; not M04 with bigger numbers.
G02	Skip-levels without undermining managers	Select: listening, follow-up, and one usable responsibility chain.
G03	Decision rights between teams	Keep for cross-team leadership: decisions and interfaces rather than reporting lines.
G04	Key-person dependence	Keep: practical transfer and tested continuity; declaring a backup is insufficient.
G05	Acquired organization integration	Specialist selection: acquisition purpose and obligations; low general preparation priority.
G06	Carrying out a reorganization	Select: transition and continuity; E04 owns the design choice.
G07	Managing former peers	Select for this transition: fairness and changed mandate; belongs near first-time management.
G08	Skeptical or exhausted people during change	Keep: understand history and capacity; do not require enthusiasm as evidence of adoption.

Talent decisions

ID	Question	Review judgment
HR01	New-hire effectiveness	Keep for managers: useful work and independence; W07 is the remote-access variant/extension.
HR02	Interviewer disagreement	Select hiring authority: evidence and resolution; M08 concerns the process.
HR03	Potential without experience	Keep for people decisions: transferable evidence and feasible support; charisma is not potential.
HR04	Promotion fairness across teams	Select calibration responsibility: opportunity access and consistent expectations.
HR05	Compensation fairness with limited budget	Select actual authority: relevant data and feasible correction; no invented formula.
HR06	Harmful high performer	Keep for managers: output does not excuse harm; K04 owns initial reporting and process.
HR07	Important person wants to leave	Keep: listen, offer honest options, respect the choice, and plan continuity.
HR08	Significant capacity reduction	Select an actual loop/role need: human consequences and commitment reset; not a routine universal question.

Technical strategy

ID	Question	Review judgment
TS01	Standards versus autonomy	Keep: shared costs and proportionate exceptions; uniformity is not inherently good.
TS02	Build, buy, or reuse	Keep: fit and whole-life obligations; avoid comparing unlike cost models.
TS03	Useful internal platform	Select platform responsibilities: consumer value, burden, adoption, and ownership.
TS04	Architecture disagreement	Keep: substantive evidence and closure; differs from defining technical quality in TS09.
TS05	Scaling bottleneck	Select technical depth: actual workload and diagnosis; fashionable architecture is not a result.
TS06	Unnecessary complexity	Keep: preserve behavior and avoid exporting burden to consumers.
TS07	Emerging technology	Keep: bounded comparison and operating implications; strong overlap with AI01 when AI is the example.
TS08	Temporary versus durable investment	Keep: horizon, real exit cost, and future ownership; can deepen a T02 episode.
TS09	What makes a technical decision good?	Added: explicit technical fit, whole-system costs, risk, and evolving context.

Culture, inclusion, and ethics

ID	Question	Review judgment
K01	Candor and high standards	Keep: practical routes and actual responses; slogans are not evidence.
K02	More inclusive participation	Keep a concrete episode: barrier, intervention, and effect; intent alone is weak.
K03	Opportunities and enabling work	Keep for work allocation: access, support, credit, and sustainability.
K04	Harmful behavior reported	Keep: responsible initial handling; not a universal legal or disciplinary script.
K05	Values versus rewards	Keep: inspect consequential incentives rather than rewriting slogans.
K06	Different working norms	Keep as a practical scenario: actual requirements versus style preferences.
K07	Questionable customer-data use	Select relevant scope: purpose, authorization, and safer alternatives with expertise.
K08	Disagreement improving a decision	Keep: decision environment and dissent; B03/B10 evaluate the candidate's particular disagreement.

Distributed work and communication

ID	Question	Review judgment
W01	Async versus meeting	Keep: purpose and channel fit; no async-only dogma.
W02	Fairness across time zones	Select distributed responsibility: real burden and decision access, not availability scoring.
W03	Hybrid proximity bias	Select hybrid work: information and opportunity patterns; overlaps K02 with a location mechanism.
W04	Useful decision record	Keep: already asks what good looks like; conclusion, status, ownership, and findability.
W05	Useful recurring meeting	Lower priority but useful: intended outcome and total cost; routine logistics alone are not enough.
W06	Conflict through remote messages	Select if relevant, or use as B03 follow-up: missing context and substantive repair.
W07	Remote newcomer context	Review possible HR01 variant: distinct value depends on distributed-access and relationship barriers.
W08	Language and access needs	Keep substantive participation expectations; do not grade accent, speed, or personal disclosure.

Personal leadership and career judgment

ID	Question	Review judgment
LQ01	Managing up amid ambiguity	Keep: clarify real conflicts and workable options; “managing up” alone is not a mechanism.
LQ02	Problematic manager decisions	Keep, with B10 overlap: work impact and bounded response, not diagnosing character.
LQ03	Sustainable work under excess demand	Review B06/B09 overlap: select the sustainability target rather than repeating prioritization.
LQ04	Seeking support and challenge	Useful optional reflection: changed thinking and retained responsibility, not name-dropping.
LQ05	Learning unfamiliar responsibilities	Keep: gaps, application, feedback, and limits; course attendance is insufficient.
LQ06	Repairing trust after a broken commitment	Keep: concrete repair and sustained behavior; an apology does not prove restoration.
LQ07	Readiness for broader responsibility	Select career/scope discussions: evidence and operating changes; tenure is not readiness.
LQ08	Consequential career tradeoff	Optional reflection with C09 overlap: make judgment understandable without grading priorities.

AI judgment and leadership

ID	Question	Review judgment
AI01	Introducing AI assistance	Select a real role need; overlaps TS07's technology adoption, with permissions and human review.
AI02	Productivity effect	Keep for AI leadership: baseline, task selection, downstream work, and inference limits.
AI03	Reviewing AI-generated code	Keep for relevant engineering work: understanding and independent proof; ownership remains human.
AI04	Learning without shaming skepticism	Select team adoption: supported capability, not compulsory enthusiasm.
AI05	AI product quality	Select product responsibility: task-specific failures and independent evaluation; no demo-only evidence.
AI06	Agent data and tool authority	Select agent responsibility: purpose, permissions, execution boundaries, and containment.
AI07	Guesses becoming facts	Keep where generated material is used; provenance and consequential reuse, with strong AI05 overlap.
AI08	Workforce and investment uncertainty	Select executive planning: scenarios and full-work effects; no vendor-uplift-to-headcount shortcut.

AI is contemporary but should not dominate a role that does not own these decisions. Forecasts need continual review; this bank establishes neither a universal productivity effect nor a staffing prediction.

Proposed next content work

These are review proposals, not additional cards already present:

• Explicitly define a good manager and good staff/principal contribution, then link those criteria to existing episodes and scenarios.
• Strengthen executive mandate, CEO/board partnership, and financial tradeoffs where the target role calls for them.
• Expand the new decision-quality illustrations into independently written historical answers and unconventional but sound alternatives. Test C07 against a disclosed range, persistent refusal, and candidate-chosen exceptions.
• Resolve flagged overlaps using actual interview loops; keep different judgment targets even when they share one evidence episode.
• Calibrate the draft anchors independently before using them for automated feedback. Do not optimize for a predetermined question count.