zora_goron

6 ms·

1.
ChatGPT Available in CarPlay(twitter.com)
3 points by zora_goron 6mo ago|0 comments
2.
Show HN: Museum Music(museummusic.samrawal.com)
2 points by zora_goron 7mo ago|0 comments
3.
Physical Intelligence: scale brings alignment between human and robot data(twitter.com)
1 points by zora_goron 9mo ago|0 comments
4.
by zora_goron 1y ago
Why might this be, does anyone know?
5.
Chrome DevTools MCP(developer.chrome.com)
1 points by zora_goron 1y ago|0 comments
6.
Claude Sonnet will ship in Xcode(developer.apple.com)
489 points by zora_goron 1y ago|400 comments
7.
Augmenting clinical reasoning, without replacing it(samrawal.substack.com)
2 points by zora_goron 1y ago|0 comments
8.
Claude Code Is All You Need(twitter.com)
4 points by zora_goron 1y ago|0 comments
9.
by zora_goron 1y ago
This difference between medical board examinations and real world practice is something that mirrors my real-world experience too, having finished med school and started residency a year ago. I’ve heard others say before that real clinical
10.
You can now connect GitHub repos to ChatGPT Deep Research(twitter.com)
2 points by zora_goron 1y ago|0 comments
11.
by zora_goron 1y ago
Anyone have any anecdotes on how expensive this is to operate, ie compared to performing the same task via Claude Code?
12.
by zora_goron 1y ago
I wrote about some similar observations in the clinical domain -- I call it the "human -> AI reasoning shunt" [0]. Explicitly requesting an AI tool to perform reasoning is one thing, but a concern I have is that, with the incre
13.
by zora_goron 2y ago
This doesn’t necessarily apply to this particular offering, but having working in clinical AI previously from a CS POV and currently from as a resident physician, something I’m a little wary of is the “shunting” of reasoning away from physi
14.
Deep research (ChatGPT) is now rolling out more widely(twitter.com)
2 points by zora_goron 2y ago|0 comments
15.
by zora_goron 2y ago
Does anyone know how this “user decides how much compute” is implemented architecturally? I assume it’s the same underlying model, so what factor pushes the model to <think> for longer or shorter? Just a prompt-time modification or so
16.
by zora_goron 2y ago
Does anyone know, how "reasoning effort" is implemented technically - does this involve differences in the pre-training, RL, or prompting phases (or all)?
17.
by zora_goron 2y ago
“A programming language is for thinking of programs, not for expressing programs you’ve already thought of. It should be a pencil, not a pen.” - from PG’s “Hackers & Painters”
18.
Contemplative LLMs(maharshi.bearblog.dev)
26 points by zora_goron 2y ago|10 comments
19.
The Human → AI Reasoning Shunt(samrawal.substack.com)
2 points by zora_goron 2y ago|0 comments
20.
How I Use Claude(borretti.me)
28 points by zora_goron 2y ago|12 comments
21.
by zora_goron 2y ago
Thanks for trying it out! Honestly, no current idea of business model - this was just something I thought would be helpful in my daily workflow so I built it out. The buttons are stored to localstorage, as is the editor text - everything st
22.
by zora_goron 2y ago
Hi there, thanks for trying it out and the kind words! These are great suggestions - thank you!
23.
Show HN: Semantic Macros Text Editor(samtxt.samrawal.com)
69 points by zora_goron 2y ago|19 comments
24.
Meta has a major opportunity to win the AI hardware race(theverge.com)
3 points by zora_goron 2y ago|1 comments
25.
Noam Brown: o1 is a "model", not a "system"(twitter.com)
2 points by zora_goron 2y ago|1 comments
26.
Show HN: Museum Music – generate period-appropriate playlists from artwork(museummusic.samrawal.com)
5 points by zora_goron 2y ago|1 comments
27.
Jeff Dean can make "reasonable approximations for complex problems quickly"(twitter.com)
4 points by zora_goron 2y ago|0 comments
28.
by zora_goron 2y ago
I'm still trying this out, but I've been using Superlist [0] (no affiliation) for different projects. For me, having a single app that is both a Markdown document <-> todo app has been helpful. I'm able to jot down note
29.
by zora_goron 2y ago
The article mentions, >> “you've tried response_format: "json" and function calling and been disappointed by the results” Can anyone share any examples of disappointments or issues with these techniques? Overall I’ve be
30.
by zora_goron 2y ago
I see a couple comments comparing llama.cpp and Ollama, and I think both have utility for different purposes. Having used both llama.cpp (which is fantastic) and Ollama, a couple things that I find valuable about Ollama out-of-the-box -- -

More ›