wg0

5 ms·

1.
by wg0 6d ago
The private equity debt time bomb is here. Then there is the bond market. And then guaranteed bust of AI bubble. 2000 billion dollars for a single company? That too when bottom of the boat has been drilled by open weight Chinese models that
2.
The Future of Software Engineering [video](youtube.com)
2 points by wg0 8d ago|0 comments
3.
by wg0 8d ago
Important - Jev is way too different, the greatest innovation are its speed and that it is guaranteed to NOT generate a token from a given set of tokens hence you can drive state machines intelligently.
4.
by wg0 8d ago
So where do I keep my pass keys is the problem. They are in my password manager just like another password. So what's the point of having them because effectively, the passwords in the password manager and the passkey in the password m
5.
by wg0 8d ago
They really are bad. Passkeys have turned out to be just another password that you cannot even memorize.
6.
by wg0 8d ago
If you have to write LAWS.bend which is pure code describing the laws then it isn't basically like those old days of writing unit tests and that too tests first hence the TDD? So what is the unique idea here except a vibe coded compile
7.
by wg0 8d ago
Some noteworthy lines from the README.md[0]: > - The compiler (not kernel) is 99% AI-written and has not been fully audited yet. > - Strings are linked lists of characters, so text processing is slow. [0]. https://github.co
8.
by wg0 9d ago
"Stealing millennium problems" would be more complete if not accurate description. And that 99.99% of the market is not interested in solving millennium problems is the other fact.
9.
by wg0 9d ago
"Slow down this much openness in AI or we won't get our trillion dollars valuations!" Google had this GPT long go and a wise man within Google noted: "We don't have any maot neither does anyone else." The AI bu
10.
by wg0 10d ago
DeepSeek is underrated. Basically all Chinese models are good enough for day to day coding at this point. The 2 trillion dollar ROI on anthropic alone? Good luck with that.
11.
by wg0 11d ago
But the real problem in self driving isn't the decision making but object description. That is, computer vision if with cameras. Decision making isn't that of a bottleneck I suppose.
12.
by wg0 11d ago
Curious what your use case is if not confidential.
13.
by wg0 11d ago
That's good. Side note - just like most people don't need an intelligent personal assistant to manage and respond their emails and book their flights, most people also don't need smart homes. Century old toggle switches are m
14.
by wg0 11d ago
Can I put it as Air Traffic Controller? With similar error rates as humans? That would be the litmus test. "Does not hallucinate" is not the same as "is never wrong". So the ATC test could be the benchmark.
15.
by wg0 11d ago
I am concerned that what will happen to detailed technical write ups that defined much of the "technical internet space"? There used to be "CodeProject" (still is around probably) that used to have great articles on C, W
16.
by wg0 12d ago
Has anyone tried it with other models.?
17.
by wg0 14d ago
Then slow down your bots too please. I deployed a website last night. 10 out of 11 GB of bandwidth went to Claude bots alone with more than 90k requests. Hypocrites hyping for the IPO. PS: True story.
18.
by wg0 17d ago
What I mean is that price has been effectively halved. During off-peak hours, the unit price is reduced from $0.007 for input cache hits to $0.003, $0.22 for input cache misses to $0.15, and $0.12 for output to $0.6
19.
by wg0 17d ago
DeepSeek v4 Flash with high is already a really great work horse. Reliable. But this time, not only that it is better but they are reducing the price by 50% so that's great. I also find the DeepSeek models to be more precise than Claud
20.
DeepSeek Flash reduced price by 100%(reddit.com)
2 points by wg0 18d ago|0 comments
21.
by wg0 1mo ago
Not happening.
22.
by wg0 1mo ago
But seriously - is software industry over? Where the next talent would come from?
23.
by wg0 2mo ago
Would love to see what was it.
24.
Fable one-shotted a Rust rewrite of the TerminalTextEffects Python library(twitter.com)
3 points by wg0 2mo ago|0 comments
25.
by wg0 2mo ago
Man - What exactly are these people?
26.
by wg0 2mo ago
Don't know about the bench marks but I am getting Opus 4.7 level performance at fraction of cost with DeepSeek V4 Flash set to high. It is a reliable workhorse.
27.
by wg0 2mo ago
I am obsessed with the idea of per tenant databases. But I am afraid of migrations. Has anyone tried that?
28.
by wg0 2mo ago
Basically: "We are in favour of 3D printers but there should be a body that tests and certifies that a 3D printer cannot print anything that can be used as weapon. Anything pointy or with a spring and recoil or... or..."
29.
GPT-5.6 found an Intel CPU microcode bug causing a performance regression(twitter.com)
2 points by wg0 2mo ago|0 comments
30.
by wg0 2mo ago
> tied to the corrupt Trump administration that are neither the highest quality nor the cheapest Someone is calling corrupt as corrupt. Surprising.

More ›