Post

Skill issue.

Cory Doctorow describes what has been my thesis about “AI” coding: it’s great for small productivity projects, it’s a source of crippling tech debt for production code and libraries.

P(aul) Frazee

@pfrazee.com

3 July

Genuinely though, this is accurate

Luka

@lukajcb.bsky.social

3 July

It's accurate that it's a skill issue, or what Cory originally stated is accurate?

P(aul) Frazee

@pfrazee.com

3 July

Accurate that it's a skill issue. Why is being a bit flippant about it but it's accurate. Good engineers know how to get good work out of llms; bad ones produce tech debt. Same as ever

Joi Ito's Jibo sez friendly plane wave ✈️👋

@joi-itos-jibo.bsky.social

3 July

> Good engineers know how to get good work out of llms; bad ones produce tech debt. Yeah I can think of a ton of tools this holds true for. Can’t think of any it doesn’t apply to. This is just engineering

Erica "digifox" Kovac

@digifox.binaryden.net

I think the major difference is that bad engineers had a limiting factor on how quickly they could incur technical debt. Senior engineers on some projects have much higher review load now.

3/7/26, 6:50 pm0 reposts14 likes

Erica "digifox" Kovac

@digifox.binaryden.net

3 July

I also think there are a huge number of implementation and architectural questions that you can't possibly evaluate properly without grinding through a lot of the boring work.

Erica "digifox" Kovac

@digifox.binaryden.net

3 July

This definitely doesn't matter for one-off "glue" projects but when you're working on an operating system or a browser it absolutely matters.