ai

The Harness of the Agent Harness

Inside the DeepSeek Harness repo is another harness: a documentation system that helps agents keep track of context, decisions, failures, and hard-won engineering lessons.

A Tale of Two Agent Harnesses

Pi and DeepSeek Harness, two philosophies for controlling complexity, preserving engineering judgment, and building agent-native software.

Less is More

As AI makes code abundant, the harder engineering judgment is deciding which code deserves to remain.

The Silence at the Top of the Hill: AI, Absurdity, and the Questions We Should Ask

The AI era confronts us with competing futures, existential anxiety, and a question Camus knew well: how do we keep searching when the world refuses to answer?

The AI-Assisted Frontend Trap: Fast Demos, Fragile Systems

Coding agents can produce impressive frontend demos quickly, but production systems still depend on explicit state, verifiable rules, and human architectural judgment.

The 40-Plus-Year AI Joke That Still Holds

A Daniel Dennett and Marvin Minsky joke still captures the paradox of fluent AI: useful enough to trust, strange enough to doubt, and powerful enough to demand human judgment.

Code Review in the Age of AI: Why It Matters More, Not Less

As AI makes code faster to produce, code review becomes the place where teams turn individual output into shared understanding, judgment, and durable software.

Rituals Against Autopilot

A reflection on rituals, intention, and staying present in the age of AI.

The Brain That Rewrites Itself: How I Did It and How You Can Too

A reflection on neuroplasticity, reinvention, and how AI can become a scaffold for thinking.

Better Without AI Than Enslaved by Its Whispers

The real danger of delusional spiral is not AI itself, but surrendering our judgment to it.

What Motivates Us in the Age of AI

A reflection on Drive, flow theory, and what it means to preserve autonomy, choice, and responsibility in the age of AI.

The Extended Mind in the Age of AI: Extension or Replacement?

Why Storytelling Is Harder Than Coding for LLMs

The Copernican View of Intelligence: We May Not Be the Center After All

The Loop We Share: What Reasoning Traces Reveal About How Humans Actually Think

The Hidden Risk of AI at Work: Faster Individuals, Fragmented Teams

After the Spark: AI and the Mind That Won’t Settle

The Cost of Letting AI Read for Us

Let’s talk about AI Psychosis/Brain Fry

MCP Is Not Dead. You’re Just Using It Wrong.

Harness Engineering: The Missing Layer Between AI Coding and Safe Automation for Financial Institutions

When AI Writes the Code, What Do You Write?

I’m Not Sure I’m a Programmer Anymore

The Love Language of AI

How to Keep Your AI-Generated Codebase Clean and Maintainable: The CRISP Strategy

Your AI Assistant Could Be Working for Hackers: The Hidden Prompt Injection Threat

Beyond Transformers: RockAI’s Alternative Path to AGI

Understanding AI’s Economic Impact Beyond Jobs

The Great NPM Shift: How AI Tools Redefined JavaScript’s Package Manager

Anthropic:AI Blackmail Is Real. Here’s What You Need to Know.

Confessions of an AI Anxious Mind: Why I’m Learning to Love the Revolution

Understanding LLM Embeddings: The Hidden Language AI Speaks

Finding Your Perfect AI Assistant: The Smart Way to Choose LLMs

Meet Gemini Diffusion: Google’s Bold New Approach to AI Writing

Is AI Cracking Reality’s Source Code?

Attention is All You Need: The Simple Idea Behind Today’s AI Revolution