you are not the sea

Nov Getting Wider Working inside other organizations as a forward deployed engineer, the AI adoption curve I see is getting wider: at the front each step makes the next one cheaper, at the back each step not taken makes the next one more expensive. Cloud and git closed that gap eventually; I don't know whether this one does. , Tunnel Vision A support document for a half-built feature came back with "not implemented yet" scattered through it. Claude and Codex both prop drill, write tests for 1+1 and over-disclose, and most of it is tunnel vision: the task in hand feels like the most important thing in the world. My partial fix is separate steps in separate contexts.

Oct The Importance of Backpressure Kanban cards were backpressure in physical form: a fixed number in circulation, so upstream could not outrun downstream. Software kept the board and dropped the pressure. , One Hundred, and Fifty I hit 100 merged PRs in a day this afternoon, not long after turning 50. Both are round numbers, and both are more useful as questions than as trophies. , Build Your Own Lightsaber Vibe code your own harness tools. Like building your own lightsaber, it teaches you how the thing actually works, and it should be cheap enough to throw away. , More Attention, Not Less I set a goal of 100 PRs a day because I was drowning in context switches. The target forced me to treat my own attention as the bottleneck. , Don't Run It Twice Most of what CI does on an agent's PR is repeat what the agent ran ten minutes earlier. So the agent leaves a git trailer listing the checks it already ran, and the PR build skips those. That was a good third of taking ten-minute builds to ninety seconds, and main re-runs everything anyway. , Start With One Tell someone an agent merged to production with no human in the loop and the reply is "that's not possible." Nobody is proposing the general case, so ask whether one is possible instead: mine was a minor patch bump. , Four to Five Days Claude sized a feature at four to five days, unprompted, then we shipped it in a few hours. It estimates in human days because everything it knows about estimating came from humans with calendars. A quirk, mostly, and a reminder that I don't size work in days anymore.

Sep Views of the Same Thing Specs, code, tests and docs are projections of one intent. Humans keep several to triangulate. The cost is drift. An agent that reads the code directly may need fewer (or one?).

Aug Rent Versus Buy A subscription makes idle capacity feel like loss. API pricing makes every call feel like spend. Same tokens, opposite behavior.

May Twenty Twenty Sprints How I structure my day in 2x20-minute sprints an hour -- a pomodoro shape that fell out of AI coding.

Apr Anthropic's definition of safety is too narrow Anthropic treats safety as a model behavior problem. Claude Code's recent month shows it's also reliability, pricing, and communication. They're where trust actually gets spent. , Resisting AI The skeptics say AI is another Kubernetes — a misapplied tech hype cycle. I think it looks more like the Internet, and probably much more. , Make it impossible, or Make it safe My dad's safety philosophy turns out to be a great framework for agentic coding: don't bubble-wrap the world, engineer one where you can take risks. , Vibe Designing Claude Design isn't vibe-coding. It's vibe-designing. And treating it like a coding tool might be holding it wrong. , Straight to Claude The moment a task or chore enters my head, it goes into Claude Code. , Context Windows for People You can't A/B test a performance review. When the feedback loop is too slow to iterate, you have to start with the right inputs -- and then try to break them before reality does. , Let It Rate With messy people data and long feedback loops, output schema isn't just an interface concern -- it's a useful steering mechanism. , Being A Better Manager in the AI Era AI is collapsing the value of coordination. If that was your primary value as a manager, you have a problem. , Vibe Coding Dev Tools Invest in your AI dev tools, but vibe code them. They will be eventually be disposable.

Mar The first PRs shipped by Customer Success. Last week our customer success team shipped their first two pull requests. It changed how we ship and how we communicate. , Battleship Prompts Same task, three different prompts. See which one gets closest. It's surprisingly effective and makes you a better prompter. , A Demographic Collapse of Developers Junior developer hiring is stalling. Meanwhile, the senior pipeline doesn't refill itself. , The Claude Dichotomy AI is creating a genuine dichotomy: teams drowning in AI-supercharged chaos, others are thriving. The difference is in the approach to transformation. , Tool Augmented Prompting Your linter, compiler, and test runner already talk to the AI. Make them say something useful. , Do as I say not as I TODO Your AI assistant holds a mirror up to your codebase. 5 ways to make sure it reflects the good parts. , No agents, no plan, ship it Why I prefer single-prompt one-shotting with AI coding tools over multi-step agents, swarms, and planning modes. , We're not vibe engineering Vibe coding is the headline. Engineering a new SDLC is the actual work. , Clear Your AI Mind Your context window is not a diary, don't forget to /clear. , Spin the Bottle Neck The bottleneck in software has moved. Teams are still optimising for the old (wrong) one. , Still choose boring technology Dan McKinley's classic advice on "choosing boring technology" is even more true in the AI era. , AI Prompts are Wishes AI prompts are wishes. The more powerful the wish, the bigger the potential for the curse. , Skills I'm ceding to AI My shortlist of coding skills I'm genuinely comfortable handing to AI. , Don't Believe the Hype AI vibe coding tools aren't truly there yet. But early adopters can see they will be. Those aren't the same thing.

Feb Faster horses and AI Productivity Not all AI productivity gains are real. A framework for telling the faster horses from the genuine improvements. , Infinite Monkeys AI coding tools need a high degree of parallelism to work. Without it, you're just watching a monkey type. , Know Your Models Software engineers routinely juggle multiple AI models. Other disciplines will be encountering this soon. , On choosing optimism? Optimism isn't the naive, easy position — it's the one that demands the most from you. , Sharpening the Axe (Branch) Using an Axe branch when iterating on your AI coding workflow. , A Daily AI Chat I'm using a single long-running AI chat each day to prioritize in the morning and summarize at the end. , Adaptive, not Compliance AI is brilliant at adaptive processes. It's a terrible fit for compliance ones. , Skilled Humans in the Loop Human-in-the-loop only works if the human has (and retains) the skills to actually be in the loop. , My Current AI Coding Workflow A snapshot of how I'm working with AI coding tools right now (Feb 2026). , Your Brain on GPS What happens to our navigation skills when we rely on GPS? And what might happen to our coding skills with AI?

Jan The 'Comment-Only' Prompt A new prompt I've found useful: Don't make any changes, just put a comment where you intend to make changes. , The Mythical Machine-Month We're moving into an era where a team of 5 engineers can outperform a team of 50. , Having a Senior Moment 2026: Wait, AI is a senior engineer now? , AI and abandonware Can we have AI maintain abandoned npm packages? , The pace of AI tools The rate that AI coding tools are adding workflow features is wild. , Return to Office (or) Agent Is RTO throwing away organizational infrastructure that AI agents need?

Earlier What is the 'why now' of your startup? What is the 'Why Now' of your startup? , Will we become gut bacteria for AI? What is the role for humanity in a world of AI? , Do you feel comfortable being flown by an AI pilot? Self-Flying. Will AI and Machine Learning replace your Captain and First Officer? , What will self-driving cars do to cities? Once we have electric and self-driving cars, what happens to all the land blighted and consumed by roads? , Building super-detailed personas? Getting all your persona assumptions down is important. You need to be wrong as much as you're right. , Will you see bots in your email? Bots in email? With a dash of JSON-LD your emails become semantic. And open up new innovations. , Don't call me and the end of no-reply? Is it finally time to end no-reply emails? , What is the Entrepreneur's Journey? Being an entrepreneur changes you. Are you working from where or who you want to be , Have you licensed your ruby gems? We're used to Ruby gems being permissive licenses, but that's not always the case. Using gemfresh you can see what licences are in your Gemfile. , Are your ruby gems fresh? Gemfresh is a quick and easy tool to see which Ruby gems in your Gemfile are ready for an update.