Outcome Bias

Latest essays

• Cursor vs Claude Code vs Codex: Only the Vibe Coder Remains — Cursor, Claude Code and Codex now look alike on paper. A dated, practical guide to picking by the shape of your work, what each costs, and how to test it.
• Running Multiple AI Agents Without Losing Your Judgment — Running several AI agents rarely costs you control in one dramatic moment. It leaks away one approval at a time. Here is how to keep judgment and avoid burnout.
• What Is AI Inference? The Using Stage Runs on Context — AI inference is the using stage: the model acts on whatever is in its window right now. Learn the four context misses that waste it, and how to fix each one.
• Don't Automate What Shouldn't Exist: Build Better AI Loops — Most AI automation scales existing confusion. Learn when to delete a process, when to use a loop, and how to build one without creating more noise.
• New AI Models Reward Systems — Opus 4.8 and Dynamic Workflows show why AI value is shifting from better answers to managed, verified work across reusable systems.
• Stop Being Loyal to One AI Stack — Claude Code, Codex app, GPT-5.5, and Opus 4.7 are not religions. Build a portable AI setup that routes work by surface, not tribe.
• 100x Your AI. Same Model as Everyone Else. — Steve Yegge says the best AI users are 100x more productive than the rest. Identical models. Different architecture. Here's the index card.
• The Wiki That Reads for You: Studying AI Behind Glass — Karpathy dropped a short note last week on building a wiki that reads for you. I built mine and pointed it at Claude Mythos — the AI Anthropic won't release.
• The Skill of Unlearning — AI doesn't make you better — it amplifies who you already are. The skill that determines which side you land on isn't learning. It's letting go.
• AI Mastering, PART IV — Your First Product Isn't the App — Same model, different operating system. A non-technical founder's guide to building the AI system that builds products — skills, hooks, agents, and pipelines.
• AI Mastering, PART III — Start with CLAUDE.md — MCPs, skills, agents — the AI tooling hype is loud. The single most impactful thing you can do takes 15 minutes: write a great CLAUDE.md. Here's how.
• Strategic Ignorance: The Skill Nobody Teaches You — The most informed people aren't consuming the most — they're ignoring the most, deliberately. Here's how to build the system.
• You Already Failed at AI. That's the Point. — Staying current with AI is a game designed to make you lose. Here's why accepting that unlocks a better question — and a system that actually works.
• Unteachable Lessons: Why Knowing Is Not Enough — Some lessons resist transfer by telling. The smartest humans across 2,500 years all noticed this — none solved it by explaining better.
• AI Mastering, PART II — AI Is Not a Chat Interface — 80% of enterprises using gen AI report no material impact. Chat makes you faster. Going beyond chat expands what you can do. A guide past the chat window.
• AI Mastering, PART I — 5 Levels of AI Chat — 800 million people use ChatGPT. 95% never leave the free tier. A framework mapping five distinct levels of AI usage — and the specific moves to climb each one.
• Stop Worshipping Effort — The effort heuristic makes you overvalue suffering and distrust efficiency. Output = Skill × Leverage / Time — and you're maximizing the wrong variable.