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 508 posts across 54 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 (98), AI Agent Architecture (61), AI Tools (38), and AI Coding Agents (27). 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.
Claude Opus 5.5 raises the capability ceiling while GPT-6 Sol and Luna cut the cost of routine agent work. Compare pricing, benchmarks, access, and the workloads each model fits.
Pat Simmons compares GPT-6 Astra, Claude Fable 5.1, and GPT-5.6 Sol across a kart racer, physics simulator, 3D product page, and website recreation.
Peter Gostev tests GPT-6 Astra across voxel London, D-Day, an open-world game, one-shot 3D scenes, reasoning levels, and direct comparisons with Claude Fable.
A source-checked comparison of Claude Fable 5.1, Gemini 3.8 Flash, and GPT-6 Astra across coding quality, task cost, speed, safeguards, and practical model routing.
Theo tested Claude Opus 5 against Fable 5 and GPT-5.6 Sol in a real T3 Code planning workflow. Here is the corrected benchmark story, cost reality, scope-control failure, and a practical routing policy.
Zubair Trabzada tested Claude Opus 5 and Fable 5 on tool use, a cinematic Higgsfield website, and a playable Three.js game. Here is the real scorecard, the test limitations, current pricing, and a better model-routing rule.
AI for Mortals tested Opus 5, Opus 4.8, and Fable 5 across web design, 3D, games, motion graphics, and knowledge work. Here is the practical scorecard, real build-cost table, and an organized prompt framework you can reuse safely.
Nate Herk tested Claude Opus 5 and Fable 5 across code, video, design, research, slides, computer use, and simulation. Here is the honest scorecard, including costs, run time, verification, and one revealing test mistake.
Nick Saraev spent about $400 testing Claude Opus 5 across interactive simulations and benchmarks. Here is what the demos prove, what remains unverified, how effort changes cost, and where Alex Finn reaches a different verdict.
Codex vs Claude Code, Cursor, and Devin on models, harness quality, included usage, and price, updated for current Opus 5 and Fable plan rules and routing.
Opus 5 nears Fable at half the price, but it stops early, argues with mature workflows, and improves when effort is lowered. The practical migration guide.
Dan Shipper and the Every team tested GPT-5.6 Sol for a month across coding, writing, design, and knowledge work. Here is where Sol wins, where Fable still leads, and how to route real tasks.
Pat Simmons tests GPT-5.6 Sol and Claude Fable 5 on the same three briefs: a shots.so-style mockup app, a viral MSCHF-style drop, and an interactive NYC learning platform.
Pat Simmons gave Fable 5 and Opus 4.8 the same prompts for an e-commerce store, 3D art museum, and strategy game. Here are the raw outcomes, costs, limitations, and model-routing lessons.
Three hands-on creator tests compare GPT-5.6 Sol with Claude Fable 5 across coding, creative builds, API reliability, speed, and cost. Here is the practical model-routing guide.
Codex is now a desktop super app with a built-in browser, annotation mode, and computer use. Claude Code is still a terminal-first local agent that writes cleaner code. Here is an honest comparison from someone who builds on both.
ELO scores, pricing and speed for 12 image models: Grok Imagine, Midjourney v7, Flux 2 Pro, GPT Image 2 and Nano Banana, plus category winners by use case.
No-code automation tools are great until they are not. Here is the honest framework for deciding when Zapier or Make.com is enough and when you need a custom AI system.
A practical, opinionated comparison of Claude and ChatGPT from an AI systems builder with Anthropic certifications who still uses GPT models regularly. When each one wins, when neither fits, and how to decide for your business.
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.