nsingh2

8 ms·

61.
by nsingh2 7mo ago
The majority of Ask/Debug mode can be reproduced using skills. For copying code references, if you're using VS Code, you can look at plugins like [1], or even make your own. Cursor's auto mode is flaky because you don't
62.
by nsingh2 7mo ago
From [1] (2022 numbers), the median creator earned around 50 Robux per year, which is ~19 cents with the current DevEx rate, and the average was 13,500 Robux. Out of ~7.5 million creators in 2022, only 11,000 qualified for cashing out. The
63.
by nsingh2 7mo ago
From what I've read online it's not necessarily a unquantized version, it seems to go through longer reasoning traces and runs multiple reasoning traces at once. Probably overkill for most tasks.
64.
by nsingh2 8mo ago
> greed is normal and expected in a free market economy OK, technically true, just like saying "water flows downhill" when someone's house is flooding. It isn't productive, the fact is well known. "The system inc
65.
by nsingh2 8mo ago
Most of the value of Claude Code comes from the model, and that's not running on your device. The Claude Code TUI itself is a front end, and should not be taking 3-4 seconds to load. That kind of loading time is around what VSCode take
66.
by nsingh2 9mo ago
This stuff is a little messy and opaque, but the performance of the same model in different harnesses depends a lot on how context is managed. The last time I tried Copilot, it performed markedly worse for similar tasks compared to Claude C
67.
by nsingh2 9mo ago
In my own testing these models sill have a different flavor to them - Opus 4.5 for software development. Works faster, and tends to write cleaner code. - GPT 5.2 xHigh for mathematical analysis, and analysis in general (e.g. code review, pl
68.
by nsingh2 9mo ago
There is an argument to be made that that companies like Walmart and Amazon operate as planned economies. They use the same cybernetic principles, real time data monitoring and feedback loops, to solve logistics and planning. These implemen
69.
by nsingh2 9mo ago
> Let freedom ring What happens if the Venezuelan people decide they want their oil profits to stay in Venezuela rather than flowing into oil company coffers? Will they have the "freedom" to choose that? Don't get me wrong
70.
by nsingh2 9mo ago
Reasoning like this is part of the reason why history keep repeating itself. Completely ignoring how previous US led decapitations turned out, and just hoping this time will be different. It should not be contentious at this point, the US o
71.
by nsingh2 11mo ago
That's quite the long esc key. An ortholinear layout would be nice too.
72.
by nsingh2 11mo ago
One quick way to estimate a lower bound is to take the number of parameters and multiply it with the bits per parameter. So a model with 7 billion parameters running with float8 types would be ~7 GB to load at a minimum . The attention mec
73.
by nsingh2 11mo ago
> Be kind. Don't be snarky. Edit out swipes [1] [1] https://news.ycombinator.com/newsguidelines.html
74.
by nsingh2 11mo ago
DINOv3 was released earlier this year: https://ai.meta.com/dinov3/ I'm not sure if the work they did with DINOv3 went into SAM3. I don't see any mention of it in the paper, though I just skimmed it.
75.
by nsingh2 11mo ago
Telling of where the boundary of competence is for these models. And to show that these models aren't doing what most expect them to be doing, i.e. not counting legs, and maybe instead inferring information based on the overall image (
76.
by nsingh2 1y ago
The other way around it seems, on `07-29-2025 23:24:56 UTC` went from 8 to 8.7 [1] [1] Table on https://www.tsunami.gov/
77.
by nsingh2 1y ago
Cosine similarity is the dot product of vectors that have been normalized to lie on the unit sphere. Normalization doesn't alter orthogonality, nor does it change the fact that most high‑dimensional vectors are (nearly) orthogonal.
78.
by nsingh2 1y ago
In general, it's the classic mistake of blaming people for societal-scale events instead of systems.
79.
by nsingh2 1y ago
Do you mean the text input box only allows one line? In that case, try pressing Alt+Enter for a new line. It's a little unintuitive; I would expect Shift+Enter for a new line.
80.
by nsingh2 1y ago
> It is completely typical, but at the same time abnormal to have tools with such poor usability. The main difference I see is that LLMs are flaky, getting better over time, but still more so than traditional tooling like debuggers. >
81.
by nsingh2 1y ago
These kinds of selfish policies backfire when they cut the ground out from under you. Trump and his cronies live in the US and are predominantly invested there, and undermining the US in the long term also undermines them and their interest
82.
by nsingh2 1y ago
> If this was such an obvious slam dunk on how to win the presidency, why didn't the Dems come up with such a candidate? The Democratic party in it's current state is ineffective & incompetent in many ways. > Managing to
83.
by nsingh2 1y ago
He’s definitely a good salesman. People already distrusted the elites and the `deep state` and he marketed himself as the outsider who’d fix things. That still isn’t sufficient evidence in his favor, given everything happening now.
84.
by nsingh2 1y ago
> As an example, I can think of one politician who is definitely a systems thinker (who is not nice, but is successful and generally outplays his opponents because of it) and ~80% of Hacker News is convinced beyond the shadow of any doub
85.
by nsingh2 1y ago
Same experience, these "background agents" are powered by models that aren't yet capable enough to handle large, tangled or legacy codebases without human guidance. So the background part ends up being functionally useless in
86.
by nsingh2 1y ago
In Cursor, there is an 'ask' mode that isn't as over-eager to make edits as the default agent mode.
87.
by nsingh2 1y ago
Better served to students and those unfamiliar with the field, but noisy to those familiar. Considering that much of mathematical work is done using pen/paper, it would be a total pain to write out huge variable names every time. Consi
88.
by nsingh2 1y ago
Whats are some use cases for these local small models, for individuals? Seems like for programming related work, the proprietary models are significantly better and that's all I really use LLMs for personally. Though I can imagine a fe
89.
by nsingh2 1y ago
One thing with Cursor is that you can use other non-anthropic models like o3 or Gemini. I've found o3 more useful for cracking localized difficult problems, and Gemini for large context refactoring.
90.
by nsingh2 1y ago
The religious absolutism you criticize is present in Israel's current leadership. Smotrich and Ben Gvir call Gaza reconquest and West Bank annexation a religious duty, claim that it's a redemption. Justifying tens of thousands of

More ›