UI Challenges | Notion

Rank	Challenge	Why this priority
1	MixBook Accessibility Audit	Already flagged as an open gap in your own search strategy; fastest to execute since it's analysis-led, not from-scratch design
2	Design System for a Real Product	Highest-leverage gap for Adobe/Canva/Pinterest-tier; Eco-Warriors proves you can build a system, this proves you can do it within real brand/legacy constraints
3	Dashboard Redesign	Closes the data-density gap; directly relevant to Pinterest (creator analytics) and Stitch Fix/Hopper (data-heavy decisions)
4	Automotive HUD / Infotainment	High differentiation, low competition (almost no designers your level have this), leverages your Bosch credibility and Detroit location
5	Micro UX Challenges (3-5 small pieces)	Cheap to produce, good for the visual/brand-forward track, but lower individual impact than the four above
6	(Optional) Combine #3 into #2	Flagged as an option below, not a separate build

2. Design System From Scratch for a Real Product

The pick: Don't invent a fictional brand — extend or rebuild a system for a product with real visual DNA. Two strong options:

Take Phia or LiveLook further: you already have their branded style guide context from real client work; build out a fuller component library (buttons, states, forms, empty states, error states) that you didn't have scope to do in the 5-day sprint
Or pick a dream-tier product's "stress point": e.g., design a dark-mode or accessibility-compliant extension of an existing brand system (Canva's UI kit, Pinterest's pin components) — framed explicitly as a speculative systems exercise, not a real client deliverable

Scope:

Audit existing inconsistencies (type scale drift, spacing inconsistency, color misuse) — 5-6 examples
Build token foundations: color, type, spacing, elevation
Build 8-10 components with documented states (default/hover/disabled/error/loading)
Show one "before" screen rebuilt entirely with the new system
Include a short rationale on scalability (how this system would hold up across platforms)

This is the piece I'd point to most directly when a Pinterest or Adobe recruiter asks "have you built a design system from the ground up."

Title: "Rebuilding Phia's Design Language for Scale" (or equivalent)

3. Dashboard / Data-Dense Interface Redesign

The pick: A creator-facing analytics dashboard fits your dream tier best (Pinterest creator stats, or a Mixbook "your projects" dashboard with usage/storage data). Avoid generic admin-panel clichés — tie it to a real user goal.

Scope:

Define the user goal precisely (e.g., "a creator wants to know which pins are driving saves this week," not "see all my data")
Audit a real or representative existing dashboard for information hierarchy problems — too much equal-weight data, poor scannability, weak empty/loading states
Redesign with a clear visual hierarchy: primary metric, secondary trends, drill-down detail