
Over drinks, a few colleagues and I kept circling a practical internal-tools problem: small workflow ideas often wait behind engineering capacity. We wondered whether AI tools let someone who understands the workflow make a rough prototype while the idea is still cheap to change or discard.
A dashboard mock, for example, makes a vague request concrete enough to argue about: people can point to what is missing and explain how they would use it. If that exposes a misunderstanding, revising the mock is cheaper than carrying the mistake into a finished tool. A lightweight intake form or a script for a recurring spreadsheet does the same job by making enough of the workflow tangible to show whether the proposed tool is useful.
A prototype like that is not something people can depend on. Work such as permissions, reliability, edge cases, and maintenance remains, and the first pass helps decide whether that investment is warranted. Someone asking for engineering time can bring a version of the workflow they have tried, along with a clearer account of what it needs to do.
Get new posts by email
Follow.it sends the emails and includes an unsubscribe link.
Related posts
Medium and Meaning

Making every part myself matters less to me when AI tools still let me express my ideas and taste.
Farming Expensive Agent Sessions

I’m using expensive agent sessions to refine the instructions that tell Codex how I want the work done.
Load Flow and Mandelbrot in the Browser

A browser-based load-flow solver and Mandelbrot explorer let me revisit power-system analysis, numerical precision, and the limits of small client-side tools.