{"id":43816573,"user":"tptacek","time":1745802614,"time_ago":"5 months ago","type":"comment","content":"

I didn't say anything about "slapdash".","comments":[{"id":43822450,"user":"quesera","time":1745853497,"time_ago":"5 months ago","type":"comment","content":"

Umm. Yeah, I think ya did. :)","comments":[{"id":43823035,"user":"tptacek","time":1745856583,"time_ago":"5 months ago","type":"comment","content":"

No.","comments":[{"id":43823279,"user":"quesera","time":1745857929,"time_ago":"5 months ago","type":"comment","content":"

You introduced the word into the thread. I quoted you.

Unless you're operating at some notational level above the literal, yes I think you did.","comments":[{"id":43823595,"user":"tptacek","time":1745859766,"time_ago":"5 months ago","type":"comment","content":"

Sorry, I was referring the the prompt, not the code.","comments":[{"id":43824045,"user":"quesera","time":1745862398,"time_ago":"5 months ago","type":"comment","content":"

I was referring to the prompt/prose as well.

The median-quality code just doesn't seem like a valuable asset en route to final product, but I guess it's a matter of process at that point.

Generative AI, as I've managed to use it, brings me to a place in the software lifecycle that I don't want to be. Median-quality code that lacks the context or polish needed to be usable. Or in some cases even parseable.

I may be missing essential details though. Smart people are getting more out of AI than I am. I'd love to see a Youtube/Twitch/etc video of someone who knows what they're doing demoing the build of a typical TODO app or similar, from paragraphs to product.","comments":[{"id":43824138,"user":"tptacek","time":1745862993,"time_ago":"5 months ago","type":"comment","content":"

Median-quality code is extraordinarily valuable. It is most of the load-bearing code people actually ship. What's almost certainly happening here is that you and I have differing definitions of "median-quality" commercial code.

I'm pretty sure that if we triangle-tested (say) a Go project from 'jerf and Gemini 2.5 Go output for the same (substantial; say, 2,000 lines) project --- not whatever Gemini's initial spew is, but a final product where Gemini is the author of 80+% of the lines --- you would not be able to pick the human code out from the LLM code.","comments":[{"id":43824326,"user":"quesera","time":1745864138,"time_ago":"5 months ago","type":"comment","content":"

This is probably true. I'm using your "median-quality" label, but that would be a generous description of the code I'm getting from LLMs.

I'm getting median-quality junior code. If you're getting median-quality commercial code, then you are speaking better LLMish than I.","comments":[{"id":43824390,"user":"tptacek","time":1745864533,"time_ago":"5 months ago","type":"comment","content":"

A couple prompt/edit "cycles" into a Cursor project, Gemini's initial output gives me better-than-junior code, but still not code I would merge. But you review that code, spot the things you don't like (missed idioms, too much repetition, weird organization) and call them out; Gemini goes and fixes them. The result of that process is code that I would merge (or that would pass a code review).

What I feel like I keep seeing is people who see that initial LLM code "proposal", don't accept it (reasonably!), and end the process right there. But that's not how coding with an LLM works.","comments":[{"id":43832507,"user":"quesera","time":1745934250,"time_ago":"5 months ago","type":"comment","content":"

I've gone many cycles deep, some of which have resulted in incremental improvements.

Probably one of my mistakes is testing it with toy challenges, like bad interview questions, instead of workaday stuff that we would normally do in a state of half-sleep.

The latter would require loading the entire project into context, and the value would be low.

My thought with the former is that it should be able to produce working versions of industry standard algorithms (bubble sort, quicksort, n digits of pi, Luhn, crc32 checksum, timezone and offset math, etc) without requiring any outside context (proprietary code) -- and perhaps erroneously, that if it fails to pull off such parlor tricks, and creates such glaring errors in the process, that it couldn't add value elsewhere either.","comments":[{"id":43849417,"user":"tptacek","time":1746039984,"time_ago":"5 months ago","type":"comment","content":"

Why are you hesitating to load all the context you need (Cursor will start from a couple starting-point files you explicitly add the context window and then go track other stuff down). It's a machine. You don't have to be nice to it.","comments":[{"id":43850947,"user":"quesera","time":1746048861,"time_ago":"5 months ago","type":"comment","content":"

Just the usual "is this service within our trust perimeter" hesitation, when it comes to sharing source code.

I expected to get better results from my potted tests, and to assemble a justification for expanding the perimeter of trust. This hasn't happened yet, but I definitely see your point.

Presumably it would also be possible to hijack Cursor's network desires and redirect to a local LLM that speaks the same protocol.","comments":[],"comments_count":0,"level":10,"url":"item?id=43850947"}],"comments_count":1,"level":9,"url":"item?id=43849417"}],"comments_count":2,"level":8,"url":"item?id=43832507"}],"comments_count":3,"level":7,"url":"item?id=43824390"}],"comments_count":4,"level":6,"url":"item?id=43824326"}],"comments_count":5,"level":5,"url":"item?id=43824138"}],"comments_count":6,"level":4,"url":"item?id=43824045"}],"comments_count":7,"level":3,"url":"item?id=43823595"}],"comments_count":8,"level":2,"url":"item?id=43823279"}],"comments_count":9,"level":1,"url":"item?id=43823035"}],"comments_count":10,"level":0,"url":"item?id=43822450"}],"comments_count":11,"url":"item?id=43816573"}

Comment by tptacek