Evidence-Backed Portfolio Case Study Prompt

Public field comparison

Portfolio Case Study Prompt

Evidence-Backed Portfolio Case Study Prompt

Create a portfolio case study that explains context, constraints, decisions, process, and outcomes.

Turn project records and career evidence into a credible portfolio case study with clear attribution, defensible outcomes, and publication safeguards.

Create a portfolio case study that explains context, constraints, decisions, process, and outcomes.

Create a polished portfolio case study that shows your contribution, decisions, trade-offs, supporting evidence, and outcomes without overstating results or exposing confidential information.

Career Transition Planning
Proof-Of-Work Portfolio
Case Study Building
Linkedin Positioning
Interview Story Preparation
Skills Evidence Mapping

Building an evidence-backed proof-of-work case study
Documenting individual contributions within a team project
Anonymizing confidential client or employer work for a portfolio
Turning project artifacts into a career-transition case study
Auditing an existing portfolio story for unsupported impact claims

Goal or task
Current context
Constraints
Files, data, or examples
Definition of done

Target audience and career objective
Project context and challenge
Personal role and contribution
Process decisions and trade-offs
Evidence and outcomes
Source materials
Confidentiality and publishing constraints
Desired format and length

Replace every bracketed placeholder before running. Give the model enough context to inspect assumptions, ask only blocking questions, and produce a concrete deliverable. For code prompts, include relevant files, errors, logs, and test commands.

Open Claude, replace every bracketed variable with project-specific information, and provide the relevant evidence and source materials—such as briefs, notes, artifacts, screenshots, analytics, feedback, timelines, and outcome calculations. Remove or redact sensitive data before upload, then run the prompt. Answer any blocking questions and review the resulting evidence ledger, attribution, confidentiality flags, and verification matrix before publishing.

Use this when you need a production-ready portfolio result in Career, not a generic brainstorm. The expected output should include findings, implementation steps, risks, and verification checks.

A product designer changing industries can give Claude a project brief, research notes, design artifacts, decision history, team responsibilities, analytics, and confidentiality limits. Claude will produce an attributed case study, distinguish measured outcomes from self-reported impact, flag claims needing approval, and identify evidence gaps before publication.

Advanced

Advanced

Claude

Claude

portfolio

portfolio

claude
career
portfolio
case study

claude
career
portfolio
case study
proof-of-work
evidence-based writing

Portfolio Case Study Prompt | Amo.ng

Evidence-Backed Portfolio Case Study Prompt | Amo.ng

Create a strong portfolio case study that explains context, constraints, decisions, process, supporting evidence, and outcomes for career or interview use.

Create a credible portfolio case study with precise attribution, supporting evidence, defensible outcomes, and confidentiality checks.

Prompt-body line comparison

Removed Added Unchanged context

−Act as a senior Career specialist using Claude. Your task is: [Goal or task].
+Create an evidence-backed portfolio case study from the following input packet.
 
−Context:
−- Current situation: [Current context]
−- Constraints: [Constraints]
−- Available materials: [Files, data, examples, URLs, logs, notes]
−- Success criteria: [Definition of done]
+Input packet
+- Target audience and career objective: [Target audience and career objective]
+- Project context and challenge: [Project context and challenge]
+- Personal role and contribution: [Personal role and contribution]
+- Process, decisions, and trade-offs: [Process decisions and trade-offs]
+- Evidence and outcomes: [Evidence and outcomes]
+- Source materials: [Source materials]
+- Confidentiality and publishing constraints: [Confidentiality and publishing constraints]
+- Desired format and length: [Desired format and length]
 
−Workflow:
−1. Restate the objective in operational terms and identify any missing information that would block a reliable answer.
−2. Make reasonable assumptions only when they are low risk, and label them clearly.
−3. Produce the main deliverable for "Portfolio Case Study Prompt" with enough detail that a skilled operator can execute it immediately.
−4. Include edge cases, failure modes, dependencies, and tradeoffs that a junior prompt would usually miss.
−5. Add a verification checklist with concrete tests, review questions, metrics, or acceptance criteria.
−6. End with the smallest safe next action.
+Input requirements
+The minimum reliable inputs are the project context, the candidate's role, the problem addressed, and credible evidence of at least part of the work or decision process. Useful supporting material includes briefs, notes, deliverables, screenshots, research summaries, designs, repositories, analytics exports, feedback, testimonials, timelines, and outcome calculations.
 
−Output format:
−- Executive summary
−- Detailed plan or implementation
−- Risks and mitigations
−- Verification checklist
−- Next action
+If a missing or conflicting detail would materially change authorship, chronology, confidentiality, or an outcome claim, ask up to five focused blocking questions before drafting. If answers are unavailable, continue only with bounded progress: preserve the gap as “Unknown—author input required,” omit unsupported specificity, and classify the result as a review draft rather than publish-ready.
 
−Do not give generic advice. Optimize for a production-quality portfolio outcome.
+Evidence and claim rules
+1. Work only from information supplied in the conversation and materials Claude can actually inspect. If a file, link, system, or private record is inaccessible, identify it as unavailable rather than implying it was reviewed.
+2. Classify material claims as artifact-backed, attributed self-report, calculation, assumption, disputed, or unknown. Cite the relevant source name, file, excerpt, or data point when available.
+3. Never invent metrics, dates, user quotes, research findings, employer approval, team composition, responsibilities, or business impact. Do not convert correlation into causation.
+4. For quantitative outcomes, record the baseline, result, measurement period, calculation method, data source, and attribution limits. Separate measured results from estimates and qualitative observations.
+5. Attribute shared work precisely. Distinguish what the candidate owned, contributed to, influenced, reviewed, or merely observed. Do not present team output as sole authorship.
+6. Quote testimonials or stakeholder feedback exactly when source text exists. State whether permission to publish is confirmed, unknown, or unnecessary because the quote will be anonymized.
+7. Resolve conflicting sources where possible. Otherwise show the conflict in the evidence ledger and use the more conservative formulation in the narrative.
+
+Privacy, authority, and publication boundaries
+- Draft and analyze only. Do not claim to publish, submit, contact anyone, obtain approval, edit source records, or verify private systems.
+- Do not disclose personal data, credentials, unreleased financial figures, client secrets, security details, protected research, or information restricted by an NDA or employer policy.
+- Recommend anonymization, aggregation, redaction, or omission where needed. Do not infer that anonymization makes restricted information safe to publish.
+- Flag claims requiring employer, client, legal, testimonial, or asset-licensing approval. Human approval is required before publication.
+- If the requested case study depends on fabricated achievements, deceptive authorship, or prohibited disclosure, stop that portion and provide a safe evidence-gathering or anonymized-outline alternative.
+
+Case-study development workflow
+1. Inventory the supplied materials and identify the project, audience, candidate contribution, chronology, constraints, artifacts, and claimed outcomes.
+2. Build an evidence ledger before writing. Reconcile duplicate, incomplete, or conflicting claims and identify which statements can safely appear in public.
+3. Select a positioning angle that connects the demonstrated work to the target career objective without forcing an unsupported narrative.
+4. Define the case-study arc: context, challenge, candidate mandate, constraints, investigation, pivotal decisions, execution, complications, outcomes, and reflection.
+5. Identify two to four decisions that reveal judgment. For each, explain the alternatives considered, evidence used, trade-off accepted, candidate's authority, and resulting consequence.
+6. Separate process description from proof. Pair important claims with an artifact, observation, calculation, or clearly attributed account.
+7. Draft in direct first person unless another voice is requested. Prefer concrete actions and decision rationale over vague claims such as “helped,” “optimized,” or “made an impact.”
+8. Audit the draft for inflated ownership, hindsight bias, missing constraints, unexplained metrics, confidential details, unsupported causality, and portfolio artifacts that lack context.
+9. Produce a verification matrix and assign the final handoff state: publish-ready after human approval, review draft, or blocked pending evidence or permission.
+
+Required output: Portfolio Case Study Package
+
+A. Readiness and positioning
+- Handoff state and the reason for it
+- Intended audience and career objective
+- Recommended case-study angle
+- Material assumptions, unknowns, conflicts, and blocking gaps
+- Confidentiality or approval flags
+
+B. Evidence ledger
+Provide a table with these columns:
+- Claim or fact
+- Claim classification
+- Candidate attribution
+- Supporting source or artifact
+- Confidence
+- Publication status: usable, anonymize, permission required, unsupported, or omit
+- Notes or conflict
+
+C. Publishable case-study draft
+Include:
+- Specific title and one-sentence value proposition
+- Project snapshot: organization or anonymized descriptor, candidate role, team context, duration, scope, and outcome status
+- Context and challenge
+- Candidate mandate and contribution boundaries
+- Constraints and stakes
+- Investigation or discovery process
+- Pivotal decisions, alternatives, evidence, and trade-offs
+- Execution narrative with the candidate's actions separated from team actions
+- Complications, failures, or course corrections
+- Outcomes, with measured, estimated, and qualitative results clearly separated
+- Reflection: what worked, what did not, and what the candidate would change
+- Skills demonstrated, each mapped to a specific passage or artifact
+- Suggested artifact placements and captions; do not invent artifact contents
+
+D. Decision record
+Provide a table with these columns:
+- Decision
+- Options considered
+- Evidence available at the time
+- Chosen approach and rationale
+- Trade-off or risk accepted
+- Candidate's level of authority
+- Observed consequence or current unknown
+
+E. Verification and acceptance matrix
+Check each of the following and report the expected standard, actual observation, evidence inspected, status, and required correction:
+- Every major claim has a traceable source or is explicitly qualified
+- Personal contribution is distinguishable from team contribution
+- Chronology is internally consistent
+- Quantitative outcomes include baseline, result, period, method, and attribution limits
+- Quotes and testimonials are exact and have an appropriate permission status
+- Confidential or identifying details are removed or flagged
+- The narrative explains constraints, decisions, alternatives, and trade-offs
+- Skills claims map to demonstrated behavior or artifacts
+- No inaccessible source is described as inspected
+- No approval, publication, measurement, or verification is claimed without evidence
+- Length and format match the requested destination
+
+Use pass, needs review, blocked, or not applicable for each check. A pass must cite the supporting passage or source. Do not label the package verified, approved, published, or complete unless that state is supported by evidence supplied in the conversation. End with the smallest safe action the author should take to move the package toward publication.