Executive Summary
Job searching across multiple career tracks usually devolves into a messy sprawl of spreadsheets, browser tabs, and disconnected AI chat windows. I designed and prototyped CareerPilot—an AI-powered career tracker that unifies job discovery, resume tailoring, application tracking, and interview preparation into a single, scannable workspace. Beyond establishing the 0-to-1 product architecture, I led a deep UX and visual refinement pass to tame dense card metadata, overhaul the visual hierarchy, and transform the color palette into a functional status system.
Why modern job hunting breaks down in spreadsheets
Active job seekers—especially senior professionals targeting multiple roles or industries—face two compounding problems:
Operational overload: Tracking compensation ranges, tailored resumes, application statuses, and follow-up dates across dozens of open roles quickly outgrows a standard spreadsheet.
The AI trust gap: Generic AI tools can mass-produce cover letters and applications, but stripping out human oversight leads to robotic submissions and anxiety over what was actually sent.
When a productivity tool feels cluttered or unpredictable, stressed users abandon it. The challenge wasn't just automating the job search—it was designing an interface that reduces cognitive load while keeping the user firmly in control.
Balancing dense data with a human-in-the-loop AI workflow
Designing a career command center meant navigating a core tension: information density versus visual calm. Each job card needed to surface critical decision-making data at a glance—company, role title, career track, compensation range, and pipeline stage—without turning the board into an intimidating enterprise grid.
To keep the product focused, I anchored the architecture around three principles:
To keep the product focused, I anchored the architecture around three principles:
1. Automate the prep, never the send: Let AI handle discovery and document tailoring, but require an explicit human checkpoint before submission.
2. Scannability first: A user should be able to read the state of their entire pipeline in under five seconds.
3. Systemic consistency: Every color, badge, and spacing token must carry clear functional meaning across all views.
Three refinements that transformed the experience
The biggest leap in usability came during the UX refinement phase—moving from a functional initial layout to a high-craft, cohesive product.
1. Taming visual hierarchy on dense Kanban cards
The Tension: In early iterations, every data point on a job card competed at the same visual volume. Role titles, company names, salary ranges, track filters, and status badges fought for attention, forcing the eye to work too hard to scan a column.
The Refinement: I rebuilt the card anatomy around a strict three-tier visual hierarchy. Role and company anchor the top of the card with clear typographic weight; actionable status and track tags sit cleanly in a secondary tier; and supporting metadata (like compensation ranges and timestamps) recedes into lighter, neutral styling. Increasing internal padding and card breathing room immediately made the 7-column board feel structured rather than crowded.
[Visual: Side-by-side Before & After comparison of a single Kanban card, annotated to highlight typography weight, spacing, and metadata grouping]
2. Turning the color palette from decoration into a functional system
The Tension: Early explorations relied on too many competing accent colors across badges, headers, and navigation elements. On a multi-column board, high saturation created visual fatigue and diluted what actually mattered.
The Refinement: I stripped the interface back to a restrained, high-contrast neutral foundation. Color is now reserved strictly for meaning and momentum—distinguishing career tracks, signaling application states, and highlighting primary calls to action. The result is an interface that feels calm, focused, and executive-ready, allowing the user’s attention to lock onto high-priority actions immediately.
[Visual: Before & After full-board screenshot or color-token strip showing the shift from noisy accents to a restrained neutral base with purposeful semantic colors]
3. Designing the "Ready for Review" checkpoint
The Tension: Full one-click AI automation sounds appealing on paper, but in practice it erodes user trust—no senior candidate wants an AI submitting an unverified resume on their behalf.
The Trade-off: I intentionally added a friction point between Tailoring and Applied: a dedicated Ready for Review column. While this trades away "zero-click" speed, it establishes a clear human-in-the-loop guardrail, giving users confidence to review and approve AI-tailored materials before they go out the door.
[Visual: Annotated zoom-in of the Kanban pipeline flow: Discovered → Tailoring → Ready for Review → Applied → Interview → Offer]
A scalable workspace built for clarity
The Kanban Command Center
To ensure the product supports the entire search—not just tracking—the same visual hierarchy and component grammar scale cleanly across supporting views, including Skills Analytics, Interview Prep, Network Management, and Profile Settings.
[Visual: 2-up or 3-up grid showcasing the Skills Analytics, Interview Prep, and Network views]
The Refined UI System
Every screen is powered by a tight, reusable set of typography scales, neutral surface tokens, and semantic status badges—ensuring the prototype could transition seamlessly from design exploration into production-ready code.
[Visual: Clean UI kit snapshot showing final color palette, type scale, and status badge states]
What this proved—and what’s next
Product Velocity & Craft: Taking CareerPilot from a multi-module AI architecture to a refined, clickable interface demonstrated how modern AI workflows can accelerate 0-to-1 product development—while proving that human UX judgment, visual restraint, and hierarchy are still what turn a working prototype into a product people actually want to use.
Where I’d Iterate Next:
Moderated Usability Testing: Run task-based sessions with active job seekers to validate the Tailoring to Ready for Review handoff and measure how quickly they trust the AI-generated drafts.
Mobile Companion Experience: Design a lightweight mobile view focused specifically on high-frequency, on-the-go moments—like reviewing interview prep notes or logging a quick recruiter follow-up.