The Garage Door: How to Build a Digital Portfolio That Shows Your Thinking

A woodworking shop with the door propped open. A glassblowing studio with a street-facing window. You see someone working and think: that person knows what they are doing. This path teaches you how to open the garage door on your own work — so that the right people see it, understand it, and remember it.

Who this is for

You have done real work. You have designed systems, solved problems, built things that other people use. But almost nobody outside your immediate circle knows about it. Your skills are locked inside one institution, one team, one conversation at a time.

This path is for you if:

You have significant professional experience but no public portfolio or visible body of work
Your best projects belong to your employer and you are not sure how to talk about them
You work in an emerging or unnamed role that does not fit standard job categories
You suspect that your positioning — not your capability — is the thing holding you back

What this path is NOT

This is not a tutorial on building a website. It is a thinking framework for deciding what to show, how to frame it, and where to put it — grounded in evidence about how hiring actually works and what makes people pay attention.

Part 1 — The 7.4-second problem

A recruiter spends 7.4 seconds on your resume.¹ They scan in an F-pattern — job titles, company names, keywords — and move on. In those 7.4 seconds, your ten years of experience, your carefully designed systems, your hard-won expertise, are invisible.

This is not because recruiters are lazy. It is because resumes are a terrible medium for communicating what you actually do. A resume is a list of claims. A portfolio is evidence.

graph LR
    A[Resume<br/>claims about work] --> B[7.4 seconds<br/>keyword scan]
    C[Portfolio<br/>evidence of work] --> D[Minutes to hours<br/>engaged evaluation]
    style A fill:#fde8e8,stroke:#e74c3c
    style C fill:#d4edda,stroke:#5cb85c

The shift toward skills-based hiring is real but uneven. 85% of employers now claim to use it, but a Harvard Business School study found fewer than 1 in 700 hires are actually affected by degree requirement removal.² The companies that genuinely adopted it — Apple, IBM, Walmart — hired 18% more non-degreed workers, who showed 10 percentage points higher retention.²

What this means: the market is moving toward “show me what you can do,” but the infrastructure for showing it is still primitive. Most people’s evidence of capability lives in their heads, their Slack history, and their employer’s intranet. A portfolio externalises it.

Meanwhile, research on personal-branding shows a direct correlation (r=0.48) between personal brand strength and perceived employability, with career satisfaction fully mediated by that perception.³ This is not about vanity. It is about being findable by the opportunities that match what you actually do.

Why this matters for you

If you have done significant work and nobody outside your organisation knows about it, you are not competing in the market where your skills have value. A portfolio does not make you better at your job. It makes your job visible to the people who would pay more for it.

Part 2 — Proof of thinking, not proof of polish

Most portfolios fail because they show the wrong thing. They show finished outputs — a polished website, a final report, a shipped product — and leave the viewer wondering: but what did you actually do?

The key insight

Hiring managers do not want to see your best work. They want to see your best thinking. The output is evidence that the thinking happened. The thinking is what they are hiring.

Nielsen Norman Group’s portfolio research confirms this: 54.5% of hiring managers rated “explaining the research process” as the most important element in a case study — above visual polish, above outcomes, above tools used.⁴

This is the distinction between proof-of-thinking and proof of work:

graph TD
    A[Proof of work] --> B[Shows what you shipped]
    A --> C[Impressive but opaque]
    A --> D[Viewer thinks:<br/>nice, but what did YOU do?]
    E[Proof of thinking] --> F[Shows how you decided]
    E --> G[Process artifacts visible]
    E --> H[Viewer thinks:<br/>I want this person on my team]
    style E fill:#4a9ede,color:#fff
    style H fill:#5cb85c,color:#fff

A proof-of-thinking portfolio includes:

The problem and constraints — not just the brief, but what made it hard
The decisions and tradeoffs — what you chose, what you rejected, and why
Process artifacts — rough sketches, decision trees, dead ends, iterations
Your specific role — what you did versus what the team did
Reflection — what you would do differently, what surprised you, what you learned

This works for any domain. A developer showing architectural decisions. A trainer showing curriculum design choices. A consultant showing how they framed a problem. An AI practitioner showing why they designed a harness one way and not another.

Go deeper

For emerging roles that don’t fit standard categories, proof-of-thinking is especially powerful — it demonstrates transferable reasoning regardless of job title.

Part 3 — The narrative frame

You have work worth showing. Now you need to frame it so a stranger understands it in under two minutes.

The STAR method — Situation, Task, Action, Result — has been the standard for behavioural interviews since DDI developed it in 1974, and approximately 73% of employers now use it.⁵ It is a useful skeleton, but for a portfolio case study it is too thin. It tells what happened; it does not show how you think.

A stronger frame adds three layers:

graph TD
    A[Context<br/>what was the situation<br/>and what made it hard?] --> B[Constraints<br/>what limited your options?]
    B --> C[Decisions<br/>what did you choose<br/>and what did you reject?]
    C --> D[Execution<br/>what did you actually build?]
    D --> E[Outcomes<br/>what happened?<br/>use numbers if possible]
    E --> F[Reflection<br/>what would you change?<br/>what surprised you?]
    style C fill:#4a9ede,color:#fff
    style F fill:#4a9ede,color:#fff

The two highlighted nodes — Decisions and Reflection — are where proof of thinking lives. Most people skip them. A case study that says “I built X and it worked” is forgettable. A case study that says “I had to choose between X and Y; I chose X because of Z; it worked, but if I did it again I would change W” is memorable and trustworthy.

The Interaction Design Foundation makes this explicit: “Your case studies have to show that you understand constraints and that you can think about how to solve a problem with these constraints in mind.”⁶

How many projects? Three to five, curated for the audience you want to reach. Ten half-documented projects are worse than three well-framed ones. Each should take a reader 2–5 minutes and leave them thinking: this person thinks clearly under pressure.

A case study structure you can steal

Title: one line naming the problem, not the deliverable

Context: 2–3 sentences. Who needed what, and why it was hard.

My role: one sentence. What was yours vs the team’s.

Constraints: bullet list. Budget, time, technical, political.

Approach: 3–5 paragraphs. The decisions, the tradeoffs, the dead ends. Include process artifacts if you have them.

Outcome: numbers where possible (“21% improvement,” not “it went well”). If the project is in progress, say so — honest incompleteness beats fabricated completion.

Reflection: what you learned, what you would change.

Part 4 — Architecture: garden or timeline?

You have projects framed as case studies. Now: where do they live?

The default answer — a personal website with a reverse-chronological project list — works, but it is not the only option, and for many knowledge workers it is not the best one.

Two models:

graph LR
    A[Timeline portfolio<br/>projects in order] --> B[Good for:<br/>career progression<br/>linear narrative]
    C[Digital garden<br/>projects as connected nodes] --> D[Good for:<br/>ongoing thinkers<br/>interdisciplinary work<br/>emerging roles]
    style A fill:#fff3cd,stroke:#f0ad4e
    style C fill:#4a9ede,color:#fff

A timeline portfolio is a list of projects, most recent first. It works for people with a clear career arc: designer → senior designer → design lead. Each project shows growth.

A digital-garden is a living graph of interconnected ideas, projects, and thinking. It is not organised by date but by relationship. Maggie Appleton catalogued the movement; Andy Matuschak’s public working notes are a canonical example.⁷ The “working with the garage door up” metaphor comes from Robin Sloan: a physical business — a glassblowing studio, a woodworking shop — that enlivens public space by making work visible.⁷

For people in emerging or interdisciplinary roles, the garden model is more powerful because it shows how your thinking connects across domains. It favours ongoing thinkers over one-time shippers.

Platform options (2025–2026):

Platform	Best for	Tradeoffs
Obsidian + Quartz	Knowledge graph, connected notes, garden model	Requires technical setup; you own everything
Astro / Hugo	Content-first static sites, fast, customisable	Requires code comfort; no graph view built in
Substack / Medium	Writing-first, built-in audience, zero setup	You don’t own the platform; limited structure
GitHub	Code portfolios, open-source work, READMEs as case studies	Technical audience only; not a general portfolio
LinkedIn	Professional visibility, discoverability, network effects	Limited formatting; algorithmic reach is volatile
Notion (published)	Quick setup, flexible, decent design	Slow, SEO-weak, not fully yours

The strongest approach combines two: a home base you own (static site or Obsidian/Quartz garden) and a distribution channel where the audience already is (LinkedIn, Substack). Write on the home base. Distribute through the channel.

The architecture question

If your work is mostly finished projects with clear outcomes, use a timeline. If your work is ongoing thinking across multiple domains, use a garden. If you are not sure, start with three case studies on any platform and decide the architecture after you have content.

Part 5 — The confidentiality problem and learning in public

Two obstacles stop knowledge workers from building portfolios. Both have solutions.

Obstacle 1: your best work belongs to your employer.

This is common and solvable. Established strategies⁸:

Ask permission — many managers agree when framed as professional development, not self-promotion
Anonymise — “Confidential Project: Federal Exam Preparation System” with logos and names stripped
Show process, not deliverables — you are sharing your decision-making, not proprietary outputs
Use percentaged outcomes — “21% improvement in pass rates” rather than absolute numbers
Tiered disclosure — public portfolio shows capability; private walkthrough shows originals under NDA in interviews
Endorsements as proxy — a LinkedIn recommendation from a manager is evidence without disclosure

The key reframe: you are not publishing the work. You are publishing your thinking about the work. The two are different and most employers will accept the distinction.

Obstacle 2: you feel you have nothing finished enough to share.

This is where learn-in-public enters. Shawn Wang coined the term in 2018 with a core thesis: the fastest way to build expertise is to document and share your learning as it happens. “The biggest beneficiary is future you. If others benefit, that’s icing.”⁹

Austin Kleon’s Show Your Work! makes the same argument from the creative side: “a blog turns flow into stock — one little blog post is nothing on its own, but published over time it becomes your life’s work.”¹⁰

graph LR
    A[Wait until<br/>work is perfect] --> B[Never publish<br/>stay invisible]
    C[Share thinking<br/>in progress] --> D[Accumulate<br/>body of work]
    D --> E[Become findable<br/>by the right people]
    style B fill:#e74c3c,color:#fff
    style E fill:#5cb85c,color:#fff

Risks of learning in public: They are real. In the startup world, “build in public” has faced backlash — oversaturation, copycats, audience capture.¹¹ But for knowledge workers (not SaaS founders), the risk profile is different. The danger is not copycats; it is premature exposure and employer sensitivity. Manage this by: sharing principles and process, not proprietary specifics; framing publicly, detailing privately; and choosing your platform deliberately.

The distinction that matters

Building in public = sharing your product’s metrics and progress (startup context, increasingly fatigued). Learning in public = sharing your thinking and growth (knowledge work context, still underused and powerful).

Part 6 — The compound effect

A portfolio is a snapshot. A body of work is a trajectory. The difference is time.

Charles Handy described the portfolio-career in The Age of Unreason (1989) — multiple concurrent roles rather than single employment, held together by a portable, demonstrable skillset.¹² He was writing about the future. It is now the present.

graph TD
    A[Month 1-3<br/>3 case studies<br/>establish presence] --> B[Month 4-8<br/>ongoing writing<br/>build the garden]
    B --> C[Month 9-12<br/>recognised voice<br/>invited to rooms]
    C --> D[Year 2+<br/>body of work<br/>career asset]
    style A fill:#fde8e8,stroke:#e74c3c
    style D fill:#4a9ede,color:#fff

The compound effect works because each piece of published thinking does three things simultaneously:

Evidence — it demonstrates your capability to anyone who finds it, at any time, without you being in the room
Clarity — the act of writing forces you to articulate what you know, exposing gaps you didn’t see (the Feynman effect)
Network — it attracts people who think about the same problems, opening doors that cold outreach never would

Personal branding research confirms the mechanism: brand differentiation — being known for a specific intersection of skills — correlates with more frequent promotions and salary increases.¹³

The practical implication: do not wait until you have a perfect portfolio to start. Publish one case study. Then another. Then a reflection piece. The garden grows. The compound interest accumulates. And three years from now, the sum of what you have published will be an asset that no employer can take away and no algorithm can suppress.

The principle

A portfolio is something you build once. A body of work is something you tend continuously. The first gets you a job. The second gets you a career.

What you now understand

Concepts you have gained

The 7.4-second problem — why resumes fail and portfolios work

proof-of-thinking — showing decisions and process, not just outputs

The narrative frame — Context → Constraints → Decisions → Execution → Outcomes → Reflection

digital-garden vs timeline — two portfolio architectures and when to use each

Confidentiality strategies — how to show institutional work without disclosure

learn-in-public — the practice of sharing thinking in progress

The compound effect — how a portfolio becomes a body of work over time

Check your understanding

Test yourself before moving on

Explain why proof of thinking is more valuable to a hiring manager than proof of polish — what specific evidence supports this?

Distinguish between a timeline portfolio and a digital garden — when is each model the better choice?

Design a case study for one of your own projects using the narrative frame: Context, Constraints, Decisions, Execution, Outcomes, Reflection

Evaluate a portfolio you admire: which of the six elements from the narrative frame does it include? Which does it skip?

Propose a strategy for documenting your most significant professional project without disclosing proprietary information

Where to go next

Build the distribution engine

You have the portfolio. Now distribute it. content-strategy teaches how to build a content engine: what to publish, how often, where, and how to sustain it without burning out.

Best for: anyone ready to start publishing consistently.

Own a platform

LinkedIn is where most professional visibility lives. If you want to understand how the algorithm works, what content formats perform, and how to build traction, read linkedin-ecosystem.

Best for: anyone whose audience is on LinkedIn.

Understand what the market wants

If you are building a portfolio for a career move into AI, ai-career-landscape maps the roles, skill stacks, salary ranges, and entry points so you can tailor your portfolio to what hiring managers actually look for.

Best for: anyone targeting a specific market.

Learn how to learn in public

If you want the evidence-based approach to self-directed learning — how to use AI without cognitive debt, the five learning loops, retrieval practice — read ai-self-learning. Then apply it publicly.

Best for: anyone who wants to learn and build their portfolio simultaneously.

Explorer

The Garage Door: How to Build a Digital Portfolio That Shows Your Thinking

The Garage Door: How to Build a Digital Portfolio That Shows Your Thinking

Who this is for

Part 1 — The 7.4-second problem

Part 2 — Proof of thinking, not proof of polish

Part 3 — The narrative frame

Part 4 — Architecture: garden or timeline?

Part 5 — The confidentiality problem and learning in public

Part 6 — The compound effect

What you now understand

Check your understanding

Where to go next

Sources

Further reading

Graph View

Table of Contents

Backlinks

Explorer

The Garage Door: How to Build a Digital Portfolio That Shows Your Thinking

The Garage Door: How to Build a Digital Portfolio That Shows Your Thinking

Who this is for

Part 1 — The 7.4-second problem

Part 2 — Proof of thinking, not proof of polish

Part 3 — The narrative frame

Part 4 — Architecture: garden or timeline?

Part 5 — The confidentiality problem and learning in public

Part 6 — The compound effect

What you now understand

Check your understanding

Where to go next

Sources

Further reading

Footnotes

Graph View

Table of Contents

Backlinks