Snorting the AGI with Claude Code • Hacker News Board

Re: Snorting the AGI with Claude Code

#251

Post by joshstrange » Wed, Jun 18, 2025, 10:11 AM UTC

TZubiri wrote: post #156

mjrbrennan wrote: post #99

Not trying to be rude here, but that `last_week.md` is horrible to me. I can't imagine having to read that let alone listen to the computer say it to me. It's so much blah blah and fluff that reads like a bad PR piece. I'd much rather scan through commits of the last week. I've found this generally with AI summaries...usually their writing style is terrible, and I feel like I cannot really trust them to get the facts…

Remember the sycophant bug? Maybe making the user FEELGOOD is part of what makes it feel smart or like a good experience. Is the reward function being smart? Is it maximizing interaction? Does it conflict with being accurate?

I ran the prompt as-is on one of the main repos that I work on and the sycophancy was cloying.

It praised so many things that I would just consider table steaks and made simple tweaks or features sound like massive projects.

I’m sure it could be improved by tweaking the prompt and there were parts of it that I found impressive that it had picked out (specifically things not in commit messages) but I found it unusable in its current form.

Re: Snorting the AGI with Claude Code

#252

Post by dghlsakjg » Wed, Jun 18, 2025, 3:58 PM UTC

TeMPOraL wrote: post #234

dghlsakjg wrote: post #221

Earlier quoted context omitted.

Conversations with Tyler Cowen, Complex Systems with patio11 are two off the top of my head that concentrate on useful information, and certainly aren't "> 90% fluff and human interaction sounds". Unless of course people talking in any capacity is human interaction sounds, in which case, yes, every podcast is > 90% human interaction sounds.

Thanks. I didn't realize 'patio11 even has a podcast, I'll definitely want to listen to that one. > Unless of course people talking in any capacity is human interaction sounds, in which case, yes, every podcast is > 90% human interaction sounds. No, I specifically mean all the thing that is not content - hellos, jokes, emoting, interrupting, exchanging filler commentary, etc. It may add character to the show, but fro…

There is a whole subcategory of wonkish podcasts, which I consider Patio11 to be the gold standard, where it is just two people having an information dense discussion. They don't tend to make it as far in the charts as the tech bro podcasts, but once you find them, they are GOLD.

Re: Snorting the AGI with Claude Code

#253

Post by cloverich » Wed, Jun 18, 2025, 6:05 PM UTC

mvieira38 wrote: post #220

CuriouslyC wrote: post #204

Earlier quoted context omitted.

Naw, the new future (technically the present for orgs that use AI intelligently) is: The AI already generated comprehensive README.md files and detailed module/function/variable (as needed) doc comments, which you could read but end up mostly being consumed by another AI, so you can just tell it what you're trying to do and ask it how you might accomplish that in the codebase, first at a conceptual level, then in cod…

What kind of stuff are you building where that is even remotely possible? I get that generating documentation works fine, but building features just isn't there yet for non-trivial apps, and don't even get me started on trying to get the agents to backtrack and change something they did

I have them backtrack all the time, including rewriting models and underlying db, then reworking from the ground up.

Another approach is Ill dictate how an api SHOULD work, or even go nuclear and write code i want to work, and tell the they must make the test pass and cant change what i wrote. They take these constraints well ime.