Every case study, system breakdown, and field note, newest first and filterable by topic. No theory, no demos. Only systems that run in production and what it took to get them there.
This is the complete archive: all 347 posts across 40 topics, newest first, filterable by category. It holds the case studies, the model and tool comparisons, and the field notes from systems I built and run. For the curated view, where the writing sits alongside the free Claude Code skills, start at the Library instead.
The archive leans in four directions: AI Search Visibility (73), AI Agent Architecture (52), AI Tools (27), and AI Coding Agents (21). Those four account for most of what I publish, because they are where most of the client questions land.
Three places to start. The AI search visibility guide is the hub for the largest cluster and links out to every spoke in it. Grok Imagine vs Midjourney is the image model comparison, rebuilt against live leaderboard data rather than left to go stale. OutreachIQ is the longest running system breakdown here, from first prototype through to a public repo.
Nick Saraev combines Nano Banana Pro, Seedance 2.0, Claude Code, and Vercel to turn a short AI video into an animated dithered hero. Here is the complete pipeline, a safer build prompt, performance limits, and deployment checklist.
Zubair Trabzada connects GPT-5.6 Sol, Codex, Higgsfield MCP, FFmpeg, and ChatGPT Sites to build cinematic product websites. Here is the source-checked workflow, original prompt, costs, limits, and production QA.
Structured data can describe a page, but it cannot repair a brand whose copy, images, video, proof, and profiles tell different stories. Use this consistency workflow.
Griffin Wooldridge shows how to connect Mobbin MCP to Claude Code so AI agents can research real shipped UI screens before designing. Here is the practical workflow for Claude, Cursor, Codex, and AI web design.
Andrew Warner and Cathryn Lavery test Paper, an agent-native design canvas controlled by Claude Code, Fable 5, Codex, and MCP. Here is what works, what still needs taste, and when Paper beats chat-only design.
Nick Saraev shows Fable 5 generating 25 highly visual websites with sub-agents, tools, deployment, and review loops. Here is the practical JQ AI SYSTEMS version: design the prompt, verify the output, and sell the human layer.
A client AI mockup can reveal taste and ambition, but it has not solved the audience, brand, feasibility, accessibility, budget, content, or conversion problem.
Brand assets and design systems are not only visual deliverables anymore. They help AI systems connect identity signals, content surfaces, and machine-readable consistency across the web.
The strongest AI-era brands are not the ones that perform machine-made effortlessness. They are the ones that use AI carefully while keeping human taste, judgment, and strategic authorship visible.
AI can speed up web production, but it can also flatten brands into the same safe average. Here is the stack I would use to keep AI-assisted websites distinctive enough to survive AI-mediated discovery.
Helpful content is no longer only a text question. In multimodal and AI-mediated discovery, visual identity discipline becomes part of whether the brand feels coherent, memorable, and trustworthy.
Rasmic's AI UI workflow is simple and useful: collect inspiration, turn it into a design system, generate tokens and components, tweak the system first, then build with AI.
Every post here is about a system that actually shipped. Book a free call and let's talk about what could ship for you.
Book Free 30-min CallAdd your name and email to unlock recording.
Thanks for your voice message. I listen to every one myself and I will get back to you within one business day.