SwtCyber

4 ms·

1.
by SwtCyber 11d ago
Give a task ownership of some state, communicate through channels and suddenly a lot of locking just disappears from the design
2.
by SwtCyber 11d ago
I think channels deserve more emphasis here too, especially because they change the architecture rather than just swapping synchronization primitives
3.
by SwtCyber 11d ago
One thing I appreciate here is treating scheduler fairness as something you spend, not something you get for free
4.
by SwtCyber 20d ago
For a programmer, deriving intervals and chords and then hearing the code produce them is a pretty effective bridge
5.
by SwtCyber 20d ago
I don't think the article needs to get you all the way from A to S to be useful. A Fourier transform doesn't tell you why a song is good either, but understanding harmonics can still improve your mental model of sound. Same with t
6.
by SwtCyber 20d ago
Yeah, this is the advice I wish showed up more often in "learn music theory" discussions
7.
by SwtCyber 20d ago
I think there's still a pedagogical niche for what the article is doing. For someone coming from programming, "here's why these conventions aren't completely arbitrary" can be the thing that gets them interested eno
8.
by SwtCyber 20d ago
I like this less as "music theory from first principles" and more as "why some of the machinery of Western music looks the way it does"
9.
by SwtCyber 24d ago
There’s a night-and-day difference between frontier and budget models, no question. But the issue isn't the tooling at all : if you put someone who doesn't know the rules of the road on a $15k carbon road bike, they're just g
10.
by SwtCyber 24d ago
Just let him take down prod on a friday night once and leave him to deal with the PagerDuty alerts solo. Cures the urge to vibe-code infrastructure real fast
11.
by SwtCyber 26d ago
Two wires and two C cells? Careful, someone is going to put a subscription layer in front of that
12.
by SwtCyber 26d ago
Every individual product is "easy", but the moment you want two ecosystems to talk to each other you become a part-time distributed systems engineer
13.
by SwtCyber 29d ago
Truly a flawless strategy: convince every government and large company on earth that depending on US-hosted AI is a geopolitical risk.
14.
by SwtCyber 29d ago
The harder problem is designing remedies that make illegal government action actually expensive enough to discourage it, without also making officials afraid to make legitimate decisions.
15.
by SwtCyber 1mo ago
To be fair, she's not wrong about feeding children's data into llms. You can trust MS policies about isolated tenants all you want , but in practice frontline staff regularly copy-paste sensitive info past all the guidelines just
16.
by SwtCyber 1mo ago
The fact that facilitators asked people to call ai human names sounds cringe ngl. Guess it's just so managers feel psychologically better about dumping their tasks on it Wonder if we name the model Bob, will it start outputting less br
17.
by SwtCyber 1mo ago
Yeah, the internal research could be far more damaging than the existence of the harmful effects themselves
18.
by SwtCyber 1mo ago
The distinction between punishment and prevention seems important here. Meta's intent might matter when deciding liability or penalties, but if a design pattern is demonstrably harmful, "we only wanted more engagement" isn&#x
19.
by SwtCyber 1mo ago
That's fair. A lot of regulation seems to work this way: first you recognize a real harm, then spend years arguing over where exactly the boundary should be
20.
by SwtCyber 1mo ago
Investors not panicking seems rational. Companies this large can survive astonishingly large fines...
21.
by SwtCyber 1mo ago
Give the agent a read-only postgres user. Boom, I just saved you a startup
22.
by SwtCyber 1mo ago
Architecture is secondary right now. How Anthropic cleaned up the datasets for training and tuned the alignment - that's the real engineering magic. OpenAI came up with GPT, but Anthropic turned it into a fully functional tool for engi
23.
by SwtCyber 1mo ago
Journalists just took whichever numbers fit the Apple narrative better
24.
by SwtCyber 1mo ago
Vercel data as proof of market dominance is hilarious. Obviously next.js pet projects are mostly generated by Claude, it's just unmatched for frontend right now. But enterprise with their rag pipelines is sitting on Azure with OpenAI,
25.
by SwtCyber 1mo ago
Maybe, although I think cement would trade one set of manufacturing problems for another
26.
by SwtCyber 1mo ago
The passive cooling is almost as clever as the filtration
27.
by SwtCyber 1mo ago
There is something very satisfying about sawdust being a feature rather than an impurity
28.
by SwtCyber 2mo ago
Disney killing the FiveThirtyEight archives is just basic s3 cost optimization, not `ai devouring humanity's memory`. A dead site just doesn't run ads
29.
by SwtCyber 2mo ago
The "100% accurate and free of hallucinations" claim should probably be replaced with something much weaker
30.
by SwtCyber 2mo ago
I think there's an underrated difference between information generation and pedagogy here. LLMs are very good at producing more explanation, yet "more explanation" is often exactly what you don't need when learning somet

More ›