Begin with evidence of skills
Let students select experience and skills without requiring a résumé upload or account.
THE TRADE-OFF
Lower setup friction means less detail for the initial recommendations.
0 → 1 PRODUCT · CAREER DISCOVERYEarly coded MVP
ApplyFlow.
Discover opportunities your major never told you to search for.
Turning transferable experience into a broader set of career possibilities.
Explore the repository (opens in a new tab)What if the problem happens before the application?
Students can organize applications and still miss entire categories of relevant work. ApplyFlow shifted from a tracking-first concept toward skills-first discovery: explain how existing experience connects to unfamiliar roles, then let the student choose a next action.
- MY CONTRIBUTION
- Problem discovery, product direction & React/TypeScript MVP
- WHO IT’S FOR
- Students exploring internship role families beyond their major
- PROJECT STAGE
- Early coded MVP
- EVIDENCE AT A GLANCE
- Working MVP · 10 educational role archetypes
The first solution was not the strongest problem.
The initial concept was an application tracker and next-action workspace. Exploratory interview narratives suggested that spreadsheets were often adequate and switching tools introduced friction.
A more useful opportunity emerged: unfamiliar job titles create a research burden. Students needed help translating experience into role families and distinguishing required qualifications from preferred ones. The direction changed before a larger tracking product was built.
10
Educational role archetypes
Fictional catalog, not active employer openings.
5
Steps in the core journey
Experience → Discover → Fit → Save → Act.
0
Required uploads or accounts
The inspected MVP stores data in the browser.
THE PRODUCT INSIGHT
A tracker helps students manage the opportunities they already know. Discovery helps them see the opportunities they were missing.
The choices that shaped the product.
Explain the connection
Show how experience relates to responsibilities and separate required from preferred qualifications.
THE TRADE-OFF
Deterministic skill overlap is transparent, but it is not a validated fit score or eligibility prediction.
Make the next action personal
Save opportunities locally and let students choose actions, notes, dates and completion status.
THE TRADE-OFF
Local storage keeps the MVP simple; data is device-specific and does not sync.
A decision you can walk through.
Explore the product thinking through a small, interactive illustration. The example content below is fictional and is not a screenshot or a live product integration.
INTERACTIVE ILLUSTRATION Fictional role examples · No live openings
My experienceDiscoverUnderstand fitSaveNext action
01 / MY EXPERIENCEStart with what
Start with what
you’ve already done.
Select example skills to reveal transparent connections. This illustration uses simple skill overlap.
No upload. No account. This case-study interaction resets when you leave.
EDUCATIONAL EXAMPLE
Product Operations Intern
Connected through Research
Turn recurring feedback into a structured set of themes.
- Required example
- Coordinate details and document findings
- Preferred example
- Exposure to product workflows
EDUCATIONAL EXAMPLE
UX Research Intern
Connected through Research
Explain what you learned from observing a user task.
- Required example
- Ask clear questions and synthesize observations
- Preferred example
- Exposure to usability testing
ORIGINAL FIGMA ARTIFACTSDesktop prototype (opens in a new tab)Mobile prototype (opens in a new tab)Original tracking concept (opens in a new tab)URLs from the repository. Figma viewing permissions may apply.
Simple enough to inspect and iterate.
Experience input
React and TypeScript capture editable skills and experience.
Role catalog
Ten fictional, educational role archetypes with sample qualifications.
Deterministic ranking
Selected-skill overlap orders roles; ties are alphabetical.
Local action plan
Browser storage persists saved roles, actions and notes; JSON export is available.
Source inspection confirms a React/TypeScript/Vite frontend. There is no live jobs feed, backend, validated AI matching model, or placement prediction. A web-search shortcut opens an ordinary search; it is not an employer integration.
What the evidence actually supports.
Implementation evidence
The repository contains the experience selection, role filters, explanations, saved roles, user-chosen actions, local persistence and export code. These establish a coded MVP, not demonstrated hiring impact.
Research boundaries
Discovery notes are exploratory and owner-supplied; participant details require reconciliation. No placement, conversion, or AI-accuracy claims are made. Independent usability and relevance evaluation remain next steps.
What I’m taking forward.
Changing the problem definition was the most valuable product decision. The pivot reduced the risk of building a more polished version of a tool students already had.
The next useful step
Test whether explanations help students identify genuinely useful unfamiliar role families. Establish a repeatable source of current opportunities before testing live-job relevance.
Sources & evidence notes
Reviewed November 10, 2026. Repository documentation is attributed to its author; implementation claims are distinguished from research findings and planned capabilities.
Current README & MVP scope (opens in a new tab)Product direction & discovery notes (opens in a new tab)Ranking and local-storage implementation (opens in a new tab)Educational role catalog (opens in a new tab)Download the content verification record