AI & UI/UX Design Blog
Latest AI tools and news for designers, plus classic UI/UX articles on research, systems, mobile, and Hong Kong product work.
Hermes Agent vs OpenClaw: Which Personal AI Assistant Fits You?
OpenClaw leads on channels and community skills; Hermes Agent leads on memory, automations, and migration from OpenClaw. A side-by-side read before you pick a personal AI gateway.
What Is Hermes Agent? Nous Research’s Self-Improving AI Assistant
Hermes Agent is an open-source, multi-channel assistant with a learning loop, messaging gateway, and cloud terminals—explained for designers and builders comparing personal agents.
How to Apply Harness Engineering: Rules, Skills, and Evals
A week-one harness checklist for Cursor-style agents: project contract, rules, skills, hooks, planning habits, and lightweight evals your team can actually maintain.
What Is Harness Engineering — and Why It Matters in 2026
The harness is the system around the model—rules, tools, and checks—that turns coding agents into reliable UI partners. A plain-language guide for designers and product teams.
How to Show AI Work in Your UX Portfolio (For Jobs and Freelance)
Recruiters and freelance clients both want the same proof in 2026: where AI helped, where you made the calls, and what shipped. A case study structure that works for hiring and client trust.
Designing Interfaces for AI Products People Actually Trust
Chat shells, confidence cues, error recovery, and when to drop users into normal UI. What changes when the product guesses instead of only computing.
UX Writing for AI Assistants: Starters, Replies, and Limits
How to write suggested prompts, in-thread replies, and refusal copy so assistants feel useful without sounding robotic or overconfident.
Claude Code in 2026: What Designers Need to Know
VS Code support, checkpoints, Sonnet 4.5, and skills. What changed in Anthropic's coding agent and how it affects design review on real repos.
AI Features vs. AI-Native Products: A UX Strategy Guide
When to add a copilot beside your current flow, when to rebuild around the model, and how lean teams avoid shipping a demo nobody returns to.
Mobile UX for AI Copilots: Small Screens, High Stakes
Thumb reach, voice, streaming replies, and interruption on iOS and Android. What changes when the copilot lives in a pocket, not on a wide monitor.
How to Usability-Test AI Features Before You Ship
Task scripts, trust probes, and what to log when answers change every run. Structured testing still works on probabilistic UI.
Cursor for Designers in 2026: Rules, Skills, and Design QA
Cursor is not a designer tool, but your team may ship UI from it anyway. Here is how to co-own rules, skills, and review so agent output matches your spec.
OpenAI Codex App and GPT-5.3-Codex: What Designers Should Watch
Desktop Codex, parallel agent threads, and GPT-5.3-Codex on long tasks. How OpenAI's coding stack shifts speed, and what that means for design QA.
Microsoft Copilot for Designers: Workspace Patterns Worth Studying
Copilot in Word, Excel, Teams, and Edge trains your users on side panels, citations, and confirm-before-action. Borrow the patterns, not the pixels.
GPT-5.5: What Changed for Designers and Builders
April 2026 release notes in plain language: latency, coding benchmarks, token use on Codex work, and what to re-test in your prompts.
ChatGPT Agent Mode: UX Lessons for Long-Running AI Tasks
Long-running agents raise the bar for progress UI, scope labels, and human confirmation. What to spec before your product promises autonomous work.
Claude Code vs. Codex vs. Cursor: A 2026 Comparison for Design Teams
Three agentic coding paths, different strengths, same design QA problem. How to partner with engineering no matter which tool wins the repo.
Mobile-First UX Design: Principles That Improve Conversion
Design for one-handed use, flaky networks, and a single primary action per screen. Mobile-first UX earns desktop complexity later.
Figma AI in 2026: Where It Helps Designers (and Where It Hurts)
Figma AI can speed up wireframes and cleanup. It can also multiply design debt if you skip tokens, tests, and handoff discipline.
Google Gemini and AI Overviews: What Product Teams Should Do
AI summaries in Search, portfolio discoverability, and people-first content. Practical steps without spammy “AI-only” pages.
Perplexity for UX Research: Fast Landscape Scans (With Verification)
Cited AI search compresses competitive and domain scans. Use it for hypotheses, then verify every source and still test with real users.
AI-Assisted User Research: 5 Rigorous Methods for Tight Timelines
Thematic coding drafts, affinity digests, and survey synthesis: five ways LLMs save hours without replacing real participants or facilitator judgment.
Anthropic Skills and MCP: Extending Claude Beyond Chat
Packaged workflows, tool connections, and agent teams. Why Anthropic's extensibility model matters for design systems and handoff.
Design Handoff When Coding Agents Ship UI All Week
Living specs, per-PR review, and what to document so Claude Code or Codex does not invent your design system on the fly.
Remote Usability Testing in One Week: A Day-by-Day Plan
Recruit five users, run think-aloud sessions, and synthesize fixes in seven days. A remote usability test plan for teams under sprint pressure.
Figma Handoff Checklist for Developers (States, Tokens, Edge Cases)
A pre-handoff checklist for states, overflow, tokens, and breakpoints so engineering and AI agents build UI without Slack archaeology.
AI News for Designers: What to Track in 2026
Themes, sources, and a filter for headlines. Stay current on agents and models without living on social feeds.
How to Build a Design System from Scratch (Without Burning Out)
Tokens first, boring components next, documentation over polish. A practical order for small teams shipping a system in 2026.
UI/UX Designer in Hong Kong: Full-Time Roles and Freelance Work
What Hong Kong companies look for when hiring in-house UI/UX—and what founders and agencies expect from freelancers. Portfolio, scope, and communication in one guide.
How Hong Kong Teams Adopt AI Coding Tools (And What Employers Look For)
Pilots, governance, and design–dev alignment on Cursor and Codex—and what Hong Kong hiring managers expect from designers on those teams.
Web vs App Design for Startups: Where to Invest First
Choose responsive web, PWA, or native app design from user habits and jobs to be done, not platform hype or investor defaults.
How to Structure a UX Case Study That Gets You Hired (or Hired as a Freelancer)
A repeatable case study outline for job applications and client pitches—context, process, decisions, and outcomes recruiters can scan in minutes.
User Research on a Budget: 5 High-Impact Methods
Five lean research methods when you have days, not months: five-user tests, hallway checks, support tickets, diary groups, and analytics.