blueboo

5 ms·

31.
by blueboo 1y ago
Do try it. The image quality and diversity is pretty shocking and not in a good way.
32.
by blueboo 1y ago
Contrast the incentives with a real tutor and those expressed in the Study Mode prompt. Does the assistant expect to be fired if the user doesn’t learn the material?
33.
by blueboo 1y ago
Or even, did any of the other successful acquisitions have potential that was perceptibly curtailed by said acquisition? Feels like a cliche at this point, but I’m not sure it’s true.
34.
by blueboo 1y ago
`llm cmd` by simonw is even better for one-off commands. It spits out the invocation and you can repurpose as suits your needs. Thanks Simon!
35.
by blueboo 1y ago
I think my TNT2 Ultra was $200. But Nvidia had dozens of competitors back then. 89 when it was founded! Now: AMD…
36.
by blueboo 1y ago
Ok. Claude Code produces most code at Anthropic. Theres an enterprise code base, with acute real needs. There are real, experienced SWEs. How much babysitting and reviewing is undetermined; but the Ants seem to tremendously prefer the workf
37.
by blueboo 1y ago
But it’s also never been easier, cheaper, simpler. So it’s not obvious that these dynamics relate to how the middle has been hollowed out
38.
by blueboo 1y ago
And I’m American, so socialism actually means western Europe in 2025!
39.
by blueboo 1y ago
Depends if socialism means the US highway system, Medicare, or The Great Leap Forward
40.
by blueboo 1y ago
Skeptics find Talking themselves out of trying them is marvellously effective for convincing themselves they’re right
41.
by blueboo 1y ago
Fred Brooks observed it in 75 As software engineers, we work with “pure thought-stuff”. We build puzzle like objects. It’s satisfying to make useful tools. It’s an ever-renewing stimulating task.
42.
by blueboo 1y ago
Retrieval and association is orders of magnitude better “Lost in the noise” no more
43.
by blueboo 1y ago
I would’ve thought the interesting angle here is about leveraging 1) the 2d space of the page and 2) color in service of compressive encoding.
44.
by blueboo 1y ago
Discourse is in the eye of the beholder. If your feed is a torrent of bile, is that evidence of a real problem in the world?
45.
by blueboo 1y ago
They can do chores that clear your day to do more challenging work. That’s immensely valuable and pretty game changing
46.
by blueboo 1y ago
Fortnite‘s dark design was the reason Epic was subject of the largest FTC gaming-related fine in history. $200m+
47.
by blueboo 1y ago
To build great AI products you need to be a fluent, deeply engaged user AND understand how they work and how to bend them to your use case beyond simple prompting. We’re in a funny moment. Right now, AI tech is so powerful and capable that
48.
by blueboo 1y ago
What keeps them from doing it? it would gross out fickle researchers working on it. X people have .. their own motivations I guess . The big labs do have evals for sensitive topics to make sure it demurs from weighing on, say, Mark Zuckerbe
49.
by blueboo 1y ago
OpenAI needs a product team hiring is hard it's a high-functioning team swimming in contemporary design and eng practices code is emerging as an important battleground OpenAI has the $$$
50.
by blueboo 1y ago
we'll eventually find that the language brain matters more for math than the math brain.
51.
by blueboo 1y ago
The old joy may be gone. But the new joy is there, if you're receptive to it
52.
by blueboo 1y ago
taps the Case Against Algebra II sign https://harpers.org/archive/2013/09/wrong-answer/
53.
by blueboo 2y ago
My experience agrees with you — which is why I tried to hedge in my statement — but meanwhile tech culture treasures its stories (perhaps just-so stories) of demanding leaders getting better performance from outrageous targets.
54.
by blueboo 2y ago
But it was at 30m users, post inflection point wrt its hockey stick Facebook knew what it was doing and the emails plainly reflect that
55.
by blueboo 2y ago
> …in mature teams with high trust, people don’t have heated arguments… This dynamic flourishes when the stakes and/or uncertainty are low enough. High stakes and high uncertainty means everyone’s pushing their intuition and their r
56.
by blueboo 2y ago
Fear of R2 looms large as well. I suspect they succumbed to the nuance collapse along the lines of “Is double checking results worth it if DeepSeek eats our lunch?”
57.
by blueboo 2y ago
Your future isn’t slipping away, but you’re updating your intuition. The path forward has always been there. The system failing is the default. Allies tip the scales. You can see the board more clearly now; play accordingly
58.
by blueboo 2y ago
It is great. But the dual use implications of this tech! …with a modest modification, it can track Person Quality of your neighborhood. How better to monitor the Culture Biome of your environment? Here’s hoping we all agree on what the defi
59.
by blueboo 2y ago
It’s not quite as bad as “no way to be found out”. There are evals that suss out contamination/training on the test set. Science means using every available means to disprove, though. Incredible claims etc
60.
by blueboo 2y ago
Systems learn by removing parts, via negativa. -Taleb

More ›