The questions worth sitting with
How Not to Be Very Powerful Broken People
A tool amplifies whoever holds it. As we pick up the most powerful tools in history, the decisive variable is not the machine but the human. Why becoming fully human is the work of this age.
Being Human in 2035
Elon University's Imagining the Digital Future Center canvassed nearly 300 global experts on what AI changes about being human. The answers were less reassuring than you'd hope.
The Seventeen Percent Problem
McKinsey's 2025 workplace AI report found a 3x gap between what leaders think employees are doing with AI and what employees are actually doing. That gap is the real crisis.
Seven of the Top Ten Skills Are Human
The WEF Future of Jobs Report 2025 said the quiet part loud: the skills that matter most in the AI era are the ones machines can't replicate.
Learn Anything Faster: The Science of Rapid Skill Acquisition
A practical playbook for rapid skill acquisition: deconstruct any skill, drill the high-leverage 20 percent, and reach useful proficiency in weeks.
The Teacher's Mindset: Education in the Age of AGI
In a world of super-intelligence, the most valuable skill isn't solving problems. It's defining them. We must all think like teachers.
The Personal Operating Layer: Your Digital Twin
Imagine an agent that knows your context, habits, and preferences. It's not just an assistant; it's your Personal Operating Layer.
Reliability in Chaos: The New Skill
In a fast-changing world, what you know matters less than what you can reliably do in a situation you've never seen before.
Talent Development Is Infrastructure
Why education needs to stop being a social service and start being treated like a power grid.
Systems Thinking: Why Understanding Connections Beats Analyzing Parts
How viewing challenges through relationships and patterns creates more effective solutions, deeper understanding, and better outcomes than traditional reductionist approaches
Structurally Unstructured: The Paradoxical Framework That Accelerates Learning
How embracing the tension between structure and flexibility creates ideal learning environments, fosters personalization at scale, and prepares individuals for an increasingly unpredictable world
How to Build Powerful Web-Search Agents That Actually Work
Nine concrete techniques for building web-research agents that are fast, cheap, and don't hallucinate — from query decomposition to adversarial cross-checking.
Cut Your Coding-Agent Bill, Part 1: Where the Money Actually Goes
Most of your coding-agent bill is grind, not genius. Part 1 of 3: how token pricing really works, why the loop bleeds you, and the plan-high, execute-cheap fix you can apply today.
Cut Your Coding-Agent Bill, Part 2: Route the Loop to a Cheaper Model
Part 2 of 3: point the coding agent you already use at a cheaper open-weight model with two environment variables. Which models hold up for execution, how to wire it, and the real economics.
Cut Your Coding-Agent Bill, Part 3: Run the Loop on Your Own Hardware
Part 3 of 3: take the agent loop fully local. Ollama, llama.cpp and vLLM, quantization and VRAM math explained, the OpenAI-compatible local endpoint, and exactly when owning the hardware beats renting tokens.
The Agent OS: How a Personal AI Agent Actually Works
A personal AI agent feels like it's thinking about you. It isn't. It's a cron job, a database, and a model. Here is the seven-layer architecture, layer by layer, with the underlying tech and the security trap.
You Grew a Mind. Now Read It: A Field Guide to Mechanistic Interpretability
A first-principles walkthrough of mechanistic interpretability: superposition, sparse autoencoders, and circuits, and how we are learning to read what a language model is actually doing.
How AI Changes Work: From Personal Tool to Org-Wide Teammate
AI at work is moving from a personal tool to a shared teammate. The three shifts rewiring how teams operate: single to multiplayer, sync to async, reactive to proactive.
Automated Development: Soon You Won't Write Code, You'll Build the Machine That Does
The job is shifting from writing code to building the pipeline that writes and ships it. A field guide to automated development: the history, the mechanism, and what comes next.
Loop Engineering in 2026, Part 2: The Outer Loop, Shared Memory, and How Loops Compound
Part 1 built a single agent loop. Part 2 is the systems layer: the outer loop that decides what to work on next, and the shared artifact store that makes independent loops learn from each other and compound into an operating system for continuous improvement.
Loop Engineering in 2026, Part 1: What an Agent Loop Is and How to Build One
The practical version of loop engineering: what an agent loop actually is, the six parts every loop is assembled from, how Claude Code's /goal works, a PR-babysitter you can build today, why iterations not tokens are the real cost, and when not to loop at all.
How to Build Powerful Web-Search Agents That Actually Work
Nine concrete techniques for building web-research agents that are fast, cheap, and don't hallucinate — from query decomposition to adversarial cross-checking.
The Founder's Stack 2026: One Person, a Team Chart of Agents
The company you used to raise a seed round to build now fits on one laptop. The solo-founder stack of 2026, layer by layer: your engineer, your back office, your marketing, your face, and how to set each up.
We Have the Power. We Don't Have the Wisdom.
Dario Amodei's 'Adolescence of Technology' names something real: we are handing civilization-scale power to systems we don't fully understand yet.
Cut Your Coding-Agent Bill, Part 1: Where the Money Actually Goes
Most of your coding-agent bill is grind, not genius. Part 1 of 3: how token pricing really works, why the loop bleeds you, and the plan-high, execute-cheap fix you can apply today.
Cut Your Coding-Agent Bill, Part 2: Route the Loop to a Cheaper Model
Part 2 of 3: point the coding agent you already use at a cheaper open-weight model with two environment variables. Which models hold up for execution, how to wire it, and the real economics.
Cut Your Coding-Agent Bill, Part 3: Run the Loop on Your Own Hardware
Part 3 of 3: take the agent loop fully local. Ollama, llama.cpp and vLLM, quantization and VRAM math explained, the OpenAI-compatible local endpoint, and exactly when owning the hardware beats renting tokens.
The Agent OS: How a Personal AI Agent Actually Works
A personal AI agent feels like it's thinking about you. It isn't. It's a cron job, a database, and a model. Here is the seven-layer architecture, layer by layer, with the underlying tech and the security trap.
Being Human in 2035
Elon University's Imagining the Digital Future Center canvassed nearly 300 global experts on what AI changes about being human. The answers were less reassuring than you'd hope.
The Seventeen Percent Problem
McKinsey's 2025 workplace AI report found a 3x gap between what leaders think employees are doing with AI and what employees are actually doing. That gap is the real crisis.
You Grew a Mind. Now Read It: A Field Guide to Mechanistic Interpretability
A first-principles walkthrough of mechanistic interpretability: superposition, sparse autoencoders, and circuits, and how we are learning to read what a language model is actually doing.