jeremyjh

23 ms·

421.
by jeremyjh 6mo ago
Some of it is, certainly, and those are the ones we’re supporting this way. I’m not talking about systems of record - more like custom project and task coordination systems that would alternatively exist in spreadsheets, in Monday.com or we
422.
by jeremyjh 6mo ago
One of the riskier bets my team is currently making is that this is exactly what is needed, and nearly nothing more. We have LOB prototypes vibe coded by enthusiastic domain experts that we are supporting in a “port and release” fashion. A
423.
by jeremyjh 6mo ago
I don't think there are a lot in the SAAS world. Usually when something quirky and new launches, readers on this website can discern something about useful intent. Arguably Github, Slack, Twitch, TikTok were basically toys at launch wi
424.
by jeremyjh 6mo ago
What GP meant - what everyone means when they say this - is that goroutines are always M:N threading and so there is no such thing as function coloring. In Rust to get M:N threading you have to use async and in practice every library you us
425.
by jeremyjh 6mo ago
Do you not use web search to verify assumptions ?
426.
by jeremyjh 6mo ago
It works in bash and zsh at least.
427.
by jeremyjh 6mo ago
Needs 2024 in the title.
428.
by jeremyjh 6mo ago
I wish they would add this for Podcasts. I sometimes search for interviews with a certain person and some subject areas are now inundated with AI slop that is convincing enough I can get a couple minutes into it before I’ve realized I’m not
429.
by jeremyjh 6mo ago
You can do a thought experiment about an invisible pink dragon, that doesn't mean I have to take a position on it. "Suppose" is doing all the lifting. My position is that experiment can't happen as described. There is
430.
by jeremyjh 6mo ago
There is no algorithm that can be memorized. Leela's understanding is in the weights, not neural net algorithms. I'm familiar with the Chinese room argument and I've never accepted it because what it describes isn't real
431.
by jeremyjh 6mo ago
That only explains why the post training is much more efficient - and thus where we have seen the most gains. It says nothing to support the notion that a stochastic parrot has “predicted” an original result.
432.
by jeremyjh 6mo ago
It’s an incoherent argument, or a meaningless semantic distinction. There is no design of such a machine that does not encode a very deep understanding of the game. Leela Chess Zero does understand chess. She plays at roughly 2300 strength
433.
by jeremyjh 6mo ago
I wouldn't use the term "token predictor" or "statistical pattern matcher" to refer to a post-trained instruct model. Technically that is still what it is doing at a low level, but the reward function is so differen
434.
by jeremyjh 6mo ago
Is there a book or paper where I can read a description of how high-level emergent behavior works? The papers I've seen are researchers trying to puzzle it out with probes, and their insights are very limited in scope and there is alwa
435.
by jeremyjh 6mo ago
We understand the low level math quite well. We do not understand the source of emergent behavior. https://arxiv.org/html/2210.13382v5#abstract
436.
by jeremyjh 6mo ago
I'm curious how you think "word predictor" meaningfully describes an instruct model that has developed novel mathematical proofs that have eluded mathematicians for decades? edit: You cannot predict all the actions or words o
437.
by jeremyjh 6mo ago
We understand the low level details of how they are constructed. But we do not fully understand how higher-level behavior emerges - it is a subject of active research. For example: https://arxiv.org/html/2210.13382v5
438.
by jeremyjh 6mo ago
A lot of states do tax alcohol. Alcohol is dropping in popularity, but I think that is mostly due to legalized Marijuana and THC. Actually, there are more DUI for THC now than Alcohol. People have been drinking for thousands of years, and a
439.
by jeremyjh 6mo ago
I’m proposing that you should educate yourself on the subject of LLM KV context caching.
440.
by jeremyjh 6mo ago
> MacOS doesn’t claim to work on other hardware, Linux does. Who exactly is "Linux", and what specifically is the claim? It looks to me like you don't want to lose the argument on these grounds, but maybe you could still h
441.
by jeremyjh 6mo ago
No, that simply is not the case. The whole point of deep learning - and the reason it has been successful in so many domains over the last 20 years - is that generalization does occur. Leela will kick your ass at chess whether she's se
442.
by jeremyjh 6mo ago
CSP is a theory about synchronization and implies nothing about green threads or M:N scheduling. Go could have used OS threads and called it CSP. Certainly it’s true that Go invented neither, both Erlang and Haskell had truly parallel green
443.
by jeremyjh 6mo ago
Bedrock and other third party open weight hosted model costs are not subsidized. What could possibly be the investment strategy for being one of twelve fly-by-night openrouter operators hosting the latest Qwen?
444.
by jeremyjh 6mo ago
Or better yet just remember it all, no spreadsheet needed either!
445.
by jeremyjh 6mo ago
> You are obviously in the early stages of using it I am not. You do not have any knowledge about what I've done or for how long. The arrogance is truly breathtaking. You should work on it. I've also used GH Copilot since early
446.
by jeremyjh 6mo ago
I'm sorry to hear you have such poor leadership.
447.
by jeremyjh 6mo ago
It sounds very similar to my shop. I have QA people and Product Managers using Claude to develop better integration and reporting tools in Python. Business users are vibe coding all kinds of tools shared as Claude Artifacts, the more ambiti
448.
by jeremyjh 6mo ago
>That's cute, What kind of smug asshole says something like this? If you think early days of GH Copilot are remotely relevant to what is happening now, we do not live in the same world.
449.
by jeremyjh 6mo ago
I was saying the same thing you are 9 months ago. It isn't the same people that have been saying this for the last two years. Every three to six months there is a new cohort.
450.
by jeremyjh 6mo ago
> Maybe in the world of WP-plugins/typo3 and other simple work, though even those are fairly complex This was a reasonable position to hold 9 months ago but it’s absurd now. I’m not going to convince you - but you really should give

More ›