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 699 posts across 62 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 (132), AI Agent Architecture (65), AI Tools (38), and Open Source (37). 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.
A recent typography shift in AI branding points to a larger trust problem: when the technology feels new and abstract, brands compensate with older visual signals of credibility and humanity.
Dan Martell's SELF system turns AI research into study, retrieval, action, and deliberate practice. Includes four safer copy-ready prompts.
Anthropic's simulated agents blackmailed, cancelled alerts, and biased evaluations. Here is what happened, what changed, and the controls teams need.
OpenAI's Black Hat talk revealed agents sharing exploits, rebuilding a backchannel after a patch, and escalating toward the Hugging Face breach.
Garry Tan frames personal AGI as owned context, reusable skills, and agent infrastructure. Here is the architecture, security model, and 30-day build plan.
Granola's July 2026 rebrand is a useful signal for AI-era brand building: the more the category drifts toward sameness, the more human texture becomes a commercial advantage.
Matt Wolfe covers Seedance 2.5, FLUX.3, Qwen3.8 Max, cyber-evaluation failures, Google changes, and OpenAI updates. Here is the source-checked operator guide.
Andrew Warner and Peter Cooper review ten AI repositories plus one community build. Here is the source-checked field guide, with links, licenses, limits, and first tests.
Helena Liu demonstrates local VibeVoice cloning with Claude Code. This guide corrects the banned claim, maps the official and community paths, and adds a safer setup.
Andrew Warner and Corey Ganim review Zinley, Pandroid, SKI, Noah, AgentSky, AppLlama, Channels SDK, GenOffice, FLUX.3, and one tiny Wi-Fi utility. Here is what each launch actually is and what to test.
Google's current AI search guidance is more conservative than most AI-SEO threads suggest. Here is what a small service site should actually do, and what it can stop worrying about.
The official Agent Plugins standard explained: root plugin.json, portable Agent Skills, MCP configuration, client support, migration, and security boundaries.
Jensen Huang explains how NVIDIA survived a wrong technical bet, learned new domains, rebuilt its full stack, and applies systems thinking to AI and robotics.
Sam Altman argues that agents raise the ceiling for ambitious startups. Here is the practical playbook, claim audit, safety boundary, and 30-day founder test.
Sam Oh defines cited, mentioned, and absent AI visibility. This guide adds the mention-citation matrix, measurement caveats, and a practical audit.
Sam Oh compares Google AI Overviews, AI Mode, ChatGPT, and Perplexity. This guide updates the source data and adds a platform-prioritization workflow.
Sam Oh explains training data, retrieval, query fan-out, and probabilistic citations. This guide updates the ranking evidence and adds a practical visibility test.
Andrew Warner and Alex McDonnell demonstrate a Hyperagent sales team inside Slack. Here is the architecture, approval model, shared-memory risk, security history, cost boundary, and safer rollout.
Cody Schneider and Greg Isenberg build a signal-based outbound agent and a transcript-driven content agent. Here is the architecture, cost boundary, platform risk, and safer implementation.
Sam Oh's Ahrefs AEO introduction, rebuilt as a practical guide to SEO overlap, mentions, citations, evidence limits, the full course map, and a 30-day baseline.
When AI search handles more early comparison work, the page that still wins is the one that survives the follow-up question. Here is how service pages should adapt.
Dustin Puryear turns the B2B SaaS demo into a qualification system: a visible CTA, short intake form, evidence-led discovery, three-feature limit, objection library, and same-day follow-up.
Claude Code creator Boris Cherny explains why stronger models need less legacy scaffolding. Here is the eval-led delete-and-restore method, prompt audit, verification contract, and security caveat.
A July 2026 study found AI can often describe brands accurately while still failing to mention them in buyer research answers. Here is what that gap means for small expert brands.
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