01 //Blog
Writing from the people doing the work: practical perspectives on technology, learning, careers and the changing shape of building software.
02 //Featured
Generative AI produces new text, images, code and audio from patterns it has learned. This guide explains how that differs from earlier AI, where it genuinely helps, and where it quietly fails.
03 //All writing
Every published article, newest first. Each entry links to its own page.
An AI agent is a model given tools, context and a goal, left to decide the order of its own steps. That freedom is the whole benefit and the whole risk.
Neither option is better in the abstract. The useful question is which parts of how you work are worth protecting, and which you would happily change to match a product someone else maintains.
Workflow automation moves the predictable parts of a process to software and leaves the judgement with people. Knowing which parts are which is most of the work.
A website tells people things. A web application lets them do things. The distinction decides your build, your budget and what happens after launch.
Choosing a technology to learn is the second decision. The first is what kind of problem you want to spend your working life on.
Most of what gets called digital transformation is digitisation: the same process, now on a screen. Transformation changes the process itself.
Useful AI adoption rarely begins with a model. It begins with a process somebody already finds tedious, and a clear idea of what a better outcome looks like.
Keep moving forward
Tell us what you're trying to learn, build or grow, and we'll point you to the part of the ecosystem that fits.
