Intermediate interview questions in Design & UX, page 12
How would you secure a moderated screen-sharing session end to end?
Cover encrypted WebRTC transport, KMS-backed storage, access controls, PII redaction, and retention policies.
How do you implement design tokens in Figma for developer handoff?
This tests design-code token architecture. Strong answer shows three-tier system—primitive, semantic, component—matching developer naming, synced via Tokens Studio or native Variables. Red flag: styles with no developer-facing names or single source of truth.
Turning a vague onboarding goal into research questions
Define success metrics, decompose the funnel, pair quantitative drop-off with qualitative why.
Framework-agnostic components across React, Vue, Angular
Web components for one core, framework wrappers for ergonomics, codegen as an alternative, with their costs.
How would you architect Figma components and styles for light and dark themes?
This tests token-based system thinking and Figma variable scoping. A strong answer maps primitives to semantic tokens, binds components to semantic variables, and switches modes at the page level.
Draft a JSON error response for an invalid authentication token
Tests your ability to standardize API error contracts with RFC 7807. A strong answer returns 401 with type, title, detail, and instance, plus an actionable fix like re-authenticating. Red flag: 403, echoing the token, or exposing stack traces.
A simple automatable metric for dev-time savings
Component coverage or cycle time, gathered automatically, compared across adoption levels.
Outline a research plan to diagnose low adoption despite positive usability tests
Tests distinguishing usability from adoption drivers. Strong answers hypothesize discoverability, motivation, and timing gaps; use behavioral triangulation, contextual inquiry, and surveys. Red flag: blaming users or redesigning before diagnosing the funnel.
How would you implement a high contrast mode theme in Figma?
Tests token architecture and non-text contrast beyond color swaps. Strong answers use Figma variable modes, cover icon inversion, focus ring specs, define image treatments. Red flag: calling it palette inversion without citing 3:1 contrast or focus specs.
Refactor this to active voice and explain why it is preferred
This tests spotting passive clauses and why active voice improves clarity. Answer: rewrite to "The script processes the data and sends it to the database," noting active voice is clearer and names the actor. Red flag: claiming passive is more professional.
Recruiting channels for a B2C usability study
Compare panels for speed versus in-product intercepts for authenticity, weigh cost, screening rigor, and bias.
Structuring color tokens for a new app
Primitive palette, semantic role tokens, optional component tokens, theming benefit.
How do you translate Figma Smart Animate into CSS or JS?
Convert easing to cubic-bezier or library curves, match ms values, animate transform and opacity, escalate layout shifts to JS.
Structure an executive-ready RFC summary for a major architectural migration
One-sentence problem and ask; options with recommendation; quantified impact, tradeoffs, and acceptance criteria.
Explain token aliasing in design systems
Primitives hold raw values, semantic tokens reference them by intent, components consume only semantic ones.
How would you track clicks on a new 'Export Data' button?
Tests prod analytics design beyond console.log. Outline: typed event schema, client validation, beacon or queue with batching, and queryable storage. Red flag: fire-and-forget tracking without retry, sampling, or privacy checks.
How do you structure Figma pages for a new project and handoff?
What it tests: systems thinking and empathy for cross-functional workflows. A strong answer names a Cover page, separate WIP and Handoff pages, and a linked component library with consistent naming.
Replace a hardcoded error message with a scalable copy system
Tests decoupling UI copy from code via key-based i18n. Move the string into a JSON namespace, wire through i18next, load dynamically, and lint against hardcoded strings. Manual find-and-replace without keys or fallback is a red flag.
Client-side versus server-side A/B testing
Client-side flickers and is bypassable; server-side assigns before response, hides logic, suits backend changes.
Designing a flexible yet usable DataTable API
Sensible defaults, compound or slot-based composition for flexibility, props for common cases.
We are hiring for this. Every open role lists the topics its interview covers, so you can prepare for the real thing rather than guessing.
See open roles