AI, in practice.
Everything I write, sorted into four topics. No fluff, no frameworks. Just what works.
Everything I write, sorted into four topics. No fluff, no frameworks. Just what works.
Recovery, sleep, and a packed calendar change how and when. They don't get a vote on whether. The assistant proposes the moved or harder version so the week still happens.
Categorising spend is solved. Renegotiating bills and moving cash without you is the new claim. The useful test is what it can do after you stop watching, and what you should never let it touch unsupervised.
My stack used to assume I was at the desk. The useful tools now keep going when I'm not. Markdown in Obsidian, synced to GitHub. A bot with its own computer. That's the filter.
AI makes it tempting to hire fewer junior developers. That might save time now and create a much bigger problem later.
AI can make parts of software delivery much faster. Clients should benefit, but that doesn't mean every project suddenly costs half as much.
We blame engineers for messy code. But tech debt is usually a symptom of decisions made (or avoided) higher up.
We've turned a normal part of growth into a disorder. Sometimes feeling out of your depth just means you're in new territory.
The AI gold rush has created a new breed of consultant. Some are brilliant. Most are winging it. Here's how to tell the difference.
Scaling a mobile team takes more than tech skills. Here's what I've learnt about leadership the hard way.