vatsachak

5 ms·

121.
by vatsachak 2mo ago
Idk... Luna is great if you generate specs before implementation. Sure a Lexus is better than a used Prius, until you include price
122.
by vatsachak 2mo ago
That's fair, that doesn't mean we don't have the technology to actually make robust websites
123.
by vatsachak 2mo ago
Allow private data to be accessed through public api
124.
by vatsachak 2mo ago
Sounds like both OpenAI and Hugging Face are incompetent
125.
by vatsachak 2mo ago
They were paper clip maximizing dawg
126.
by vatsachak 2mo ago
If your code is complex enough for Luna Max to fail maybe you need to write a bit yourself so they can copy your idea
127.
by vatsachak 2mo ago
I think anthropic is behind but Luna on a Jalapeno seems profitable
128.
by vatsachak 2mo ago
Yeah browsing every stressful piece of news on X was unironically increasing my cortisol levels
129.
by vatsachak 2mo ago
I think there can be many complaints about large scale AI providers but given that large companies are now pushing for local models it seems that we are luckily avoiding the situation where AI is a utility
130.
by vatsachak 2mo ago
As a former lecturer, there's way more concerning blights than preparing founders. For example, the abuse of the disability system, excessive student loans, the lack of standardized testing at higher level courses, the use of LLMs as a
131.
by vatsachak 2mo ago
Hey I work in the casino industry and we have regulations!
132.
by vatsachak 2mo ago
It's almost like an EV should be regular car with a battery
133.
by vatsachak 2mo ago
What's the point of agents.md if you just use an LLM on a codebase? Just say, complete this bit like how the rest is... Even then they aren't great at it. Idk, the best case use for LLMs are extremely specific requests, for exampl
134.
by vatsachak 2mo ago
I hate to be brusque but this is cope. GPT 5.6 Sol is just as good and cheaper
135.
by vatsachak 2mo ago
> reinvently Slop
136.
by vatsachak 2mo ago
But I can use custom DSLs using macros in Rust or template Haskell. In Lean 4 you can even add grammar to the Lean parser. I feel like the good parts of lisps have been adopted by typed languages...
137.
by vatsachak 2mo ago
They plan to be able to run 1T parameter models next year https://www.mythic.ai/vanguard Seems big, IF true
138.
by vatsachak 2mo ago
Like the numbers they claim could literally make LLMs 20x profitable. If it were true then why isn't every AI company trying to buy them out?
139.
by vatsachak 2mo ago
If they can't demonstrate it publicly it's probably fake.
140.
by vatsachak 2mo ago
Why not just give the model a prompt? Every gain in LLMs is either through increases in compute efficiency, Architecture or Harnesses... The rest seems like bells and whistles
141.
by vatsachak 2mo ago
I've never seen the appeal in schemes other than hot reloadability...
142.
by vatsachak 2mo ago
So...emacs
143.
by vatsachak 2mo ago
I do not have the stillness required for intricate work. I remember spilling acid all over myself in the chem labs in college. Luckily for us brutes we have math and programming
144.
by vatsachak 2mo ago
All those folks telling me to update my dependencies, this is why I don't do it. It's not laziness, it's undeniable foresight.
145.
by vatsachak 2mo ago
You just proved that AI cannot currently do that
146.
by vatsachak 2mo ago
Well one person can use at most one bicycle at a time. One person can use as many GPUs as they want.
147.
by vatsachak 2mo ago
LLMs love duplicating logic. Unfortunately they are not formal systems so we actually have to read their output
148.
by vatsachak 2mo ago
It's just that flock is insecure and is also used in a negative manner by those who wield it. I mean functionally, given cell phones and ring doorbells we're already in a surveillance state.
149.
by vatsachak 2mo ago
LLMs are enormously good 1. and 2. But humans are much better at the rest. I don't want to read another LLM documentation ever again. The information to text ratio is way too small.
150.
by vatsachak 2mo ago
There's only three kinds of optimization realizations 1) Realize that your problem isn't as general as it was 2) Realize that your problem has a better memory model 3) Realize that your problem can be parallelized further Hyperopt

More ›