The "pattern setting" strategy

Simon Willison, co-creator of the acclaimed Django project, has quite a bit of experience developing software.

In several ways. He not only has a long track record as a traditional software engineer. He has also become a thought leader for agentic AI coding, writing extensively about it on his blog.

So I paid attention when he spoke about it in an interview:

"One of the magic tricks about [coding agents] is they're incredibly consistent. If you've got a codebase with a bunch of patterns in, they will follow those patterns almost to a tee...

There's a lot to be said for keeping your codebase high quality, because the agent will then add to it in a high quality way."

This suggests what I call the "pattern setting" approach.

That's when you ensure your codebase has a high-quality architecture, structure and design, demonstrating the patterns and practices you want coding agents to follow when implementing new features.

You can establish this by hand-writing the important parts of the code; or by using a coding agent.

Though you may need to write your prompts especially detailed and thorough at first, to get the high level codebase quality you need; and you better review the code carefully, to ensure it's doing what you want it to do.

Either way, then you can relax a bit.

At that point, you don't have to write a fat technical book in every prompt. The coding agent will just follow how your codebase does things already, when writing new code for you.

To pull this off requires both the old ways, and the new.

Both of which are taught in Boosted Code. Read about it below.