Re sources

‍Most software waits to be used. The most useful kind pays attention even when nobody asked it to.

The busywork that never shows up on a timesheet is quietly costing firms more than they think.

The debate about AI taking over design misses what's actually on the table: not the design, but everything around it.

Most predictions about AI in architecture focus on generating buildings. The more likely future is quieter than that.

Design tools are powerful and completely passive. That combination quietly shapes how much of the work falls on you.

The AI rollouts that stick aren't the ambitious ones. They're the narrow ones that quietly proved themselves first.

The fear isn't automation itself. It's automation nobody can see, question, or reverse.

Not everything that can be automated should be. The useful question isn't capability — it's what happens when the automation is wrong.

Trust in an AI tool isn't granted once. It's built one correct suggestion at a time — and lost much faster than it's earned.

The most important design decision in any AI tool isn't what it can do. It's what it's allowed to do without asking first.

Most firms have a family library. Far fewer have one people actually reach for instead of building from scratch.

Speed and quality are usually framed as a trade-off. In most design workflows, they're actually competing for the same limited resource: attention.

A standards document is only as useful as the last time someone actually checked their work against it.

Most coordination problems aren't caused by bad modelling. They're caused by coordination happening too late to matter.

The mistakes that cost the most time are rarely the dramatic ones. They're the small, repeatable habits that quietly compound across a project.