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 781 posts across 63 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 (137), AI Agent Architecture (72), AI Business Ideas (41), and Open Source (41). 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.
Kacper Rutkiewicz tests Scrapling across targeted web extraction, prospect research, and design synthesis. Here is the source-checked setup, the real token math, the fetcher decision tree, and the legal and security boundaries his workflows need.
Vaibhav Sisinty builds a reusable specialist, a delegated startup team, and a scheduled analyst in Hyperagent. Here is the practical no-code setup with prompts, permissions, costs, rubrics, and approval gates.
Riley Brown and Vince Canger show how to build a practical Buzz agent team with Codex as the lead, Claude Code and other runtimes as specialists, scoped channels, review loops, mobile access, and safer recurring workflows.
Andrew Warner and Vince Canger test Buzz across agent runtimes, Wasp and Railway deployment, mobile access, Git projects, workflow triggers, shared compute, and multi-agent routing. Here is the practical operator guide.
Vaibhav Sisinty tests ChatGPT Work and Claude Cowork across projects, scheduled tasks, connectors, recorded skills, Slack, Sites, Artifacts, and local work. Here is the current, source-checked comparison.
Greg Isenberg and Vinny tour Buzz, Block's open agent workspace. The real opportunity is portable team context across Claude Code, Codex, Goose, Git worktrees, workflows, and local compute, with important early-preview limits.
Cody Schneider and Greg Isenberg map a real Meta Ads agent: unified business data, research and creative pipelines, bounded API actions, and a feedback loop tied to revenue.
Run Gemma 4 in LM Studio on a 16GB integrated-graphics PC. Compare all five model sizes, official Q4 memory estimates, downloads, speed, context, vision, coding, and Bionic.
Ryan Carson explains how he runs parallel cloud agents, recurring production checks, self-improvement loops, and model routing. This guide turns the interview into a controlled agent-operations system.
Block's Buzz puts people, Claude Code, Codex, and other agents in one Nostr workspace. Here is the architecture, privacy reality, chief-of-staff pattern, shared-compute risk, and honest Slack comparison.
Ari Meisel shows how Tasklet agents coordinate website optimization, knowledge, content, support, and personal tasks. Here is the architecture, full toolkit, shadow-mode rollout, and the controls the demo still needs.
Run GPT-5.6 Sol inside Hermes Agent: the verified Codex OAuth setup, how to pick reasoning effort, three bounded workflows, and the safety controls to keep.
Vaibhav Sisinty explains why token consumption is becoming a business cost problem. Here is the source-checked playbook: measure cost per successful task, route models, cache correctly, trim context, set limits, and decide when local inference pays.
Greg Isenberg and Elie Steinbock explain how loop engineering can improve SEO, ads, and product feedback. Here is the practical architecture: objective metrics, memory, budgets, verification, schedules, and stop conditions.
Riley Brown builds a Jarvis-style desktop agent in Cursor with GPT-Realtime-2, Exa search, computer control, Mermaid artifacts, and voice-edited thumbnails. Here is the architecture, setup, cost, and safer permission model.
Dan Shipper shows how he uses GPT-5.6 and Codex as a work operating system. Here is the practical architecture behind Tend, inbox cards, pulse threads, Proof, goals, skills, and human approval.
Vaibhav Sisinty and OpenAI's Wulfie Bain use GPT-5.6, ChatGPT Work, and Codex to research, build, test, automate, and publish a startup workflow.
Greg Isenberg and Nick Vasilescu use Grok 4.5, Hermes, Orgo, memory, connectors, and review gates to move from startup research to a landing page and outreach draft in one working session.
OpenAI introduced ChatGPT Work, an agentic work surface for apps, files, local projects, browser use, Computer Use, Sites, Appshots, scheduled tasks, and GPT-5.6-powered deliverables.
Nate Herk gave Claude Fable 5 one goal prompt and got a product, landing page, videos, business plan, market research, and brand package. Here is the practical builder lesson: goal design, subagents, verification, and human validation.
Nate Herk shows how to turn YouTube transcripts, PDFs, URLs, meetings, and notes into a cross-linked LLM wiki that Claude Code, Fable 5, Codex, Hermes, and other agents can reason over.
Riley Brown connects Claude Fable 5 returning, OpenAI GPT-5.6 entering limited preview, Sonnet 5, Cursor for iOS, model rerouting, and the shift from selling apps to selling agents. Here is the practical builder read.
Ras Mic breaks down how he thinks about system design when building AI-generated apps. Here is the practical JQ AI SYSTEMS version: monorepos, client-server boundaries, durable workflows, service isolation, observability, auth, payments, and agent-friendly infrastructure.
Ras Mic and The Calum Johnson Show walk through the fastest practical way to use AI agents in business and content: connect the right tools, coach one useful run, turn it into a skill, and automate the repeat.
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.
Your recording is emailed to me and kept on this site's server for 12 months, then deleted. Privacy policy.
Thanks for your voice message. I listen to every one myself and I will get back to you within one business day.
Explore João's services and public work, or arrange a free 30-minute consultation.
Choose a time with Cal.com. Your booking status is shown there. No calendar details are shared with this assistant.
Open booking in a new tabWhen enabled, each question is processed by Cloudflare AI and checked for abuse with Turnstile. No chat history is stored by this app. Please leave out private information.
Prepared answers · Public information