News for hackers. Stuff that matters.
Re:In 27,280 coding agent runs, 64% of inference spend went ... by rrr_oh_man on Tuesday November 17, 2026 @11:19AM (#50533650)
> objectively nicer than I would've gotten from the effort I would've put into manually coding

I find that only works with topics I know very little about.

Re:In 27,280 coding agent runs, 64% of inference spend went ... by datsci_est_2015 on Tuesday November 17, 2026 @11:29AM (#50533849)
This comment could be interpreted at least two ways, and I’m not sure which it is:

  - tongue in cheek: in topics you know very little about, you don’t have the ability to tell whether the LLM output is trash or not
  - very serious: LLMs allow you to create things you could not have created before due to not having the knowledge
Sort of like a Dunning-Kruger or Gell-Mann amnesia.
Re:In 27,280 coding agent runs, 64% of inference spend went ... by devmor on Tuesday November 17, 2026 @11:38AM (#50534004)
I don't see why both can't be true.

LLMs are allowing people to create things with very little effort compared to before, and those creations are of questionable quality - indeterminable by the LLM-user due to a lack of experience.

Re:In 27,280 coding agent runs, 64% of inference spend went ... by ygjb on Tuesday November 17, 2026 @11:46AM (#50534184)
So, basically, the same as before, but LLMs provide the creations of questionable quality at a faster rate :P

I think the quality of the results vary with the rigor of the process. If you one-shot prompt a new service (even with Fable5) "build me an airtable competitor, no bugs", you will get some real code camp quality results, but a well defined set of prompts and a workflow that leverages multiple coding agents reviewing each others work, and implementing test cases at a relatively granular level yields radically different results.

Couple that with a rigorous design/spec session, and multiple checkpoints (say at release milestone) to monitor drift from the original design makes the process a bit more intentional and measurable, while still being far faster than a typical engineer can build.

Re:In 27,280 coding agent runs, 64% of inference spend went ... by yugoslavia4ever on Tuesday November 17, 2026 @11:55AM (#50534357)
Yes, but before you learned as you went, researching and finding out what good (or at least better) looked like; what learning is done when you wave your hand?
Re:In 27,280 coding agent runs, 64% of inference spend went ... by ygjb on Tuesday November 17, 2026 @10:29PM (#50541493)
Honestly? I have more time to learn now than I did before. I can also do something that was really challenging before LLMs as well, which is turn my lack of comprehension into an demand, guided tour through a Socratic exploration of the unknown topic, including generating labs specific to the skills I want to develop and with the resources I have available.

It's like anything else with LLMs, if you treat it like a magic wand, you haven't really learned or built anything, but using it as a true assistant, it can be a powerful and flexible tool.