Comment by nickdothutton

> It’s just mimicking what people say on the internet when they get performance gains

probably read bunch of junior/mid level resumes saying they optimized 90% of company by 80%

Both "better coffee" and "faster code" are measurable targets. Somewhat vaguely defined, but nobody is stopping the Barista or Claude from asking clarifying questions.

If I gave a human this task I would expect them to transform the vague goal into measurable metrics, confirm that the metrics match customer (==my) expectations then measure their improvements on these metrics.

This kind of stuff is a major topic for MBAs, but it's really not beyond what you could expect from a programmer or a barista. If I ask you for a better coffee, what you deliver should be better on some metric you can name, otherwise it's simply not better. Bonus points if it's better in a way I care about

I assume a good barista would ask some follow up questions before making the coffee.

I was experimenting with Claude Code and requested something more CPU efficient in a very small project, there were a few avenues to explore, I was interested to see what path it would take. It turned out that it seized upon something which wasn't consuming much CPU anyway and was difficult to optimise further. I learned that I'd have to be more explicit in future and direct an analysis phase and probably kick-in a few strategies for performance optimisation which it could then explore. The refund request was an amusement. It was $15 well spent on my own learning.

I could also argue if a barista gets multiple complaints about their coffee it's very much their and their employer's job to go away and figure out to make good coffee.

It's very much not the customers job to learn about coffee and to direct them how to make a quality basic coffee

And it's not rocket science.

"Optimize this code for performance" is not an unstructured or vague request.

Any "performance" axis could have been used: Number of db hits, memory pressure, cpu usage, whatever.

The LLM chose (or whatever) to use CPU performance, claimed a specific figure, and that figure was demonstrably not real.

If you ask a barista to make you a better coffee, and the barista says "this coffee is hotter" and it just isn't, the problem is not underspecified requirements, the problem is that it just doesn't make any attempt to say things that are only correct. Technically it can't make any attempt.

If I tell an intern "Optimize this app for performance" and they come back having reduced the memory footprint by half, but that didn't actually matter because the app was never memory constrained, I could hem and haw about not giving clear instructions, but I could also use that as a teachable moment to help the budding engineer learn how to figure out what matters when given that kind of leeway, to still have impact.

If they instead come back and say "I cut memory usage in half" and then you have them run the app and it has the exact same memory usage, you don't think about not giving clear enough instructions, because you should be asking the intern "Why are you lying to my face?" and "Why are you confidently telling me something you did not verify?".