batshit_beaver

6 ms·

31.
by batshit_beaver 6mo ago
This is going to be interesting. At least for coding, there's little correlation between token spend and the quality (and impact) of the resulting AI suggestion. This is fine when inference prices are capped (eg via a monthly subscript
32.
by batshit_beaver 6mo ago
My experience is that if I manage to describe a problem in enough detail for a junior or LLM to be able to solve it, it would have been faster to do it myself. Prior to LLMs the idea was to involve juniors in the engineering process to give
33.
by batshit_beaver 6mo ago
> investment in polish, quality, and reliability For there to be any trust in the above, the tool needs to behave predictably day to day. It shouldn't be possible to open your laptop and find that Claude suddenly has an IQ 50 point
34.
by batshit_beaver 6mo ago
What I want to know is why my bedrock-backed Claude gets dumber along with commercial users. Surely they're not touching the bedrock model itself. Only thing I can think of is that updates to the harness are the main cause of performan
35.
by batshit_beaver 6mo ago
Anything is possible with a big enough bank account.
36.
by batshit_beaver 6mo ago
> We "literally" couldn't design programs to do the interesting things LLMs can do. That's a bit of an overstatement. The entire field of ML is aimed at problems where deterministic code would work just fine, but the
37.
by batshit_beaver 6mo ago
First you have to figure out HOW to measure productivity.
38.
by batshit_beaver 6mo ago
You mean the company where engineers ask chat bots to write chess games in their spare time in order to hit their AI usage requirements? That Meta?
39.
by batshit_beaver 6mo ago
That might require thinking instead of feeling.
40.
by batshit_beaver 6mo ago
I'd trust that dude over professional leetcoders any day. But you're right that trust is a complicated thing and often misplaced. I think as an industry we're always reevaluating our relationship with OSS, and I'm sure L
41.
by batshit_beaver 6mo ago
Because you trust that your dependencies are not vibe coded and have been reviewed by humans.
42.
by batshit_beaver 6mo ago
You literally cannot, since ANY changes to code tend to introduce unintended (or at least not explicitly requested) new behaviors.
43.
by batshit_beaver 7mo ago
> I'm all for AI and it's great for things like copywriting, brainstorming and code generation It's funny how the assumption is always that LLMs are very useful in an industry other than your own.
44.
by batshit_beaver 7mo ago
Are you also an LLM? Do objects often begin rotating when you're only holding them with one hand?
45.
by batshit_beaver 7mo ago
That's because AI labs keep stamping out the widely known failures. I assume without actually retraining the main model, but with some small classifier that detects the known meme questions and injects correct answer in the context. Bu
46.
by batshit_beaver 7mo ago
He didn't say he won.
47.
by batshit_beaver 7mo ago
Legalese already exists in software engineering. Several dialects of it, in fact. We call them programming languages.
48.
by batshit_beaver 7mo ago
https://www.citadelsecurities.com/news-and-insights/2026-glo...
49.
by batshit_beaver 7mo ago
In theory, there's no difference between theory and practice. In practice...
50.
by batshit_beaver 8mo ago
Yeah, this. Every conversation inevitably ends with "you're absolutely right!" The number of "you're absolutely right"s per session is roughly how I measure model performance (inverse correlation).
51.
by batshit_beaver 8mo ago
But carrying bunker busters via drones is a lot more expensive than hand grenades.
52.
by batshit_beaver 8mo ago
Yeah, akin to talking to a rubber ducky
53.
by batshit_beaver 8mo ago
> This is no different then carpentry. Yes, all furniture can now be built by machines. Some people still choose to build it by hand. Does that make them less productive? Yes. I take issue even with this part. First of all, all furniture
54.
by batshit_beaver 8mo ago
This seems bizarre. Only reason my family bought a Tesla is thanks to the ev tax credit. Without it there are far better options.
55.
by batshit_beaver 8mo ago
What's interesting is that the strength of US dollar vs other currencies is barely budging in the meantime. Seems like everyone else is inflating away their problems too, so it all evens out in the end (unless you're poor with no
56.
by batshit_beaver 8mo ago
That's what the third shoe is for - aircraft carriers.
57.
by batshit_beaver 9mo ago
What's stopping you from writing code by hand even today? I mainly use LLMs for researching and trying possible paths forward, but usually implement the suggested solution myself specifically so that I fully understand the code (unless
58.
by batshit_beaver 9mo ago
Unfortunately IDEs are not yet directly connected to our minds, so there's still that silly little step of encoding your ideas in a way that can be compiled into binary. Playing the broken telephone game with an LLM is not always the m
59.
by batshit_beaver 1y ago
You use LLMs to _discover_ how to approach important problems. You don't necessarily need to use the output verbatim. Same as StackOverflow and Google.
60.
by batshit_beaver 1y ago
Min-maxing is fine, but the endless balance patches force you to relearn the game every few months. So now you have to commit to a live service game full time, just to keep up. Or you can choose to play StarCraft or chess or something where

More ›