Writing

Early Error Handling Gives Away an Agent Built the Feature in the Wrong Order

August 8, 2026 Early Error Handling Gives Away an Agent Built the Feature in the Wrong Order I've reviewed thousands of coding-agent changes. Early error handling often gives aw...

What AI Coding Benchmarks Don’t Measure

July 21, 2026 What AI Coding Benchmarks Don’t Measure The real engineering test starts with the next change. Most AI coding benchmarks stop at the one before it. The headline sc...

The Capability Is Real. The Discipline Is Sold Separately.

July 19, 2026 The Capability Is Real. The Discipline Is Sold Separately. The same demo is everywhere right now: an agent running for hours, untouched, shipping a whole feature o...

I'm Learning Go the Way I Learned Python — By Shipping It

July 16, 2026 I'm Learning Go the Way I Learned Python — By Shipping It The risk with agentic coding isn't bad code. It's a generation of engineers shipping code they don't unde...

The Agent Wrote a Better Pull Request Than the Humans

July 15, 2026 The Agent Wrote a Better Pull Request Than the Humans Everyone's braced for agents to ship code fast and sloppy. I just reviewed one that shipped fast and better (...

AI Solved "Invent." Nobody Solved "Simplify."

July 13, 2026 AI Solved "Invent." Nobody Solved "Simplify." Amazon has a leadership principle for this: invent and simplify. 37signals has a name for the second half: the wind t...

Green Was Never the Reason You Hit Merge

July 10, 2026 Green Was Never the Reason You Hit Merge "All checks have passed. This branch has no conflicts with the base branch." That green box is now the trigger some teams ...

Can Pair Programming Answer the AI PR Review Bottleneck?

July 3, 2026 Can Pair Programming Answer the AI PR Review Bottleneck? Can pair programming embed peer review directly into the development workflow? And could that be an answer ...

AI Made Me Language-Agnostic. Ruby on Rails Was Just the Proving Ground

July 1, 2026 AI Made Me Language-Agnostic. Ruby on Rails Was Just the Proving Ground For over a decade, Ruby on Rails was the only stack I fully trusted myself in as a solo or s...

The Adapter Pattern Is Commercial Leverage, Not Clean Code

June 30, 2026 The Adapter Pattern Is Commercial Leverage, Not Clean Code The adapter pattern is one of the first design patterns every engineer and agent should understand — and...

"Write Clean Code" Is a Hope, Not an Instruction

June 26, 2026 "Write Clean Code" Is a Hope, Not an Instruction Use TDD. Refactor carefully. Follow SOLID. Use design patterns. Write clean code. I've seen agent instruction file...

Your AI Code Review Tool Runs Too Late

June 23, 2026 Your AI Code Review Tool Runs Too Late AI code review tools like CodeRabbit and Greptile are useful. But if the plan is to let the agent build it and then let the ...

Stop Prompting. Start Operating.

June 21, 2026 Stop Prompting. Start Operating. The people getting the most out of AI agents are not the best prompters. They are the best operators. Keep the human at the start ...

AI Code Review: Stop Adding Reviewers, Start Fixing the Input

June 17, 2026 AI Code Review: Stop Adding Reviewers, Start Fixing the Input The fix for AI code review isn't more reviewers. It's making the code worth reviewing before it ever ...

Spec-First AI Is Just Waterfall With Better Typing

June 16, 2026 Spec-First AI Is Just Waterfall With Better Typing Spec-first AI development is quietly rebuilding the one thing engineers spent a decade escaping: waterfall. The ...

Your Next Customer Might Not Be Human — Is Your Business Ready to Sell to Them?

June 11, 2026 Your Next Customer Might Not Be Human — Is Your Business Ready to Sell to Them? Your next customer might not be human. On parts of the internet automated traffic a...

Governed PRs: How to Keep the Quality Bar at Agentic Speed

June 9, 2026 Governed PRs: How to Keep the Quality Bar at Agentic Speed AI can write more pull requests than your team can confidently review. Governed PRs are how you keep both...

AI Code Review Asks the Wrong Question

June 8, 2026 AI Code Review Asks the Wrong Question AI code review asks whether the code looks okay. Governed delivery asks whether the change is acceptable. That distinction ma...

The Token Spend Reversal: How Heavy Spend Went From Virtue to Waste

June 4, 2026 The Token Spend Reversal: How Heavy Spend Went From Virtue to Waste A year ago the message to engineers using AI was: use the agents fully, lean in, let them run. T...

AI's immediate value is not automation. It is amplification.

June 2, 2026 AI's immediate value is not automation. It is amplification. AI's immediate value is not automation. It is amplification. Most businesses have not figured that out ...

Governance has a bad reputation in software, usually for good reason.

June 1, 2026 Governance has a bad reputation in software, usually for good reason. Too often it means more process, more forms, more meetings, more gates, and more delay. Red ta...

AI doesn't remove engineering bottlenecks. It moves them and makes them expensive faster.

May 22, 2026 AI doesn't remove engineering bottlenecks. It moves them and makes them expensive faster. Old delivery had a slow feedback loop. Friction built up over months. Slow...

AI-assisted delivery changes the cost record for software engineering

May 21, 2026 AI-assisted delivery changes the cost record for software engineering AI-assisted software delivery changes how teams measure the cost of engineering work. That sou...

I am not dropping twenty years of delivery practice for agentic speed.

May 19, 2026 I am not dropping twenty years of delivery practice for agentic speed. That is the thought that kept coming back as I tried to work out what AI coding agents should...

Before You Trust Auto-Merge

May 16, 2026 Before You Trust Auto-Merge More engineering teams are talking openly about agents auto-merging PRs. The reaction is predictable — half excitement, half horror. The...

I didn't want to change good software delivery to suit agents

May 12, 2026 I didn't want to change good software delivery to suit agents Your team is shipping faster. Your review queue is longer. That’s not a coincidence. This is the patte...

Why stripe_subscription_id.present? is a governance time bomb

May 11, 2026 Why stripe_subscription_id.present? is a governance time bomb You have written this line. So have I. def can_access? stripe_subscription_id.present? endIt looked fi...

Spec Kit Is Free. Waterfall Was Too.

May 9, 2026 Spec Kit Is Free. Waterfall Was Too. Specs are useful. They are not governance. Spec Kit exists because vibe coding made an old software problem suddenly more danger...

Engineering Had Governance First. Then Marketing Demanded It Too.

May 1, 2026 Engineering Had Governance First. Then Marketing Demanded It Too. Once I had the engineering side of the Dark Factory model working, governed pull requests, agentic ...

When you own the full SDLC alone, governance is not process. It is survival.

April 29, 2026 When you own the full SDLC alone, governance is not process. It is survival. Governance is not something I learned from a book. It is something I could not afford...