mvkel

6 ms·

31.
by mvkel 3mo ago
All of these complaints are about humans making user-hostile decisions, not AI. And software quality is not a reflection of its code quality. Never has been.
32.
by mvkel 3mo ago
> This was just before Claude Code was released so this wasn't full agentic hands-off engineering. Instead, I broke down the development into lots of small tasks and went through them manually with the AI, testing each change as I w
33.
by mvkel 3mo ago
So OpenAI calling Anthropic out on their fear based marketing is an unqualified statement? Have -any- of Anthropic's concerns of imminent disaster passed muster?
34.
by mvkel 3mo ago
The most important prompt lesson everyone should take away: never ask yes / no / leading questions. Almost all of Tao's questions begin with what, or why. That forces an open ended response, which is a great way to reduce or
35.
by mvkel 3mo ago
I was simply responding to the point that Anthropic has some altruistic bent and are trying to downplay the fear, when in fact their entire commercial strategy is to convince society that they alone are qualified to hold the keys. Being in
36.
by mvkel 3mo ago
> do their nonsense to get headlines They know what they're doing. It's a playbook. You write scary stuff in the model card to make it look like legitimate whitepaper rEsEarCh, then drip-feed it to the media outlets who make it
37.
by mvkel 3mo ago
Define "winning." Open source is cheap, yet its operating systems are the least-popular. But their existence is critical to a healthy market. It's not zero sum. What you're describing is how things become commoditized, b
38.
by mvkel 3mo ago
As someone who just used computer use to successfully perform a "change of address" across 52 different services, hard disagree. It's magic.
39.
by mvkel 3mo ago
> 5 days ago: 7M users > 4 days ago: 8M users > 3 days ago: 9M users That's some incredible growth.
40.
by mvkel 3mo ago
No way. It's simply another layer of abstraction. You don't code in binary, or assembly, or C, or vanilla JavaScript anymore. Eventually "coding" will simply be prompting skill with the wisdom of architectural decisions.
41.
by mvkel 3mo ago
> Consoles made sense as a product category where specialized graphics hardware was not generally available for consumer PCs. This has almost never been true. GPUs existed, and were being used, before the N64. Your comment also begs the
42.
by mvkel 3mo ago
One tangible thing this doesn't touch on: SpeechAnalyzer supports streaming, so you can see what it's hearing from you as you talk. A massive UX improvement. Many of the other models force you to record, then it transcribes the au
43.
by mvkel 3mo ago
In my experience, the better long-term choice if you're going to vibe code an app is to use Expo. Its basis is React, so the code output quality is much higher than Swift because there is much more React code in LLM training data. Ever
44.
by mvkel 3mo ago
> "People just submitted it. I don't know why. They 'trust me'. Dumb f*ks" - Mark Zuckerberg Sometimes there are no consequences
45.
by mvkel 3mo ago
Hence the fool's errand that is "alignment"
46.
by mvkel 3mo ago
> It's interesting that all models seems to be unbiased out of the box so far This is really begging the question. If something relies on the perception of a human, it has bias. The data (or lack thereof) used to train models is per
47.
by mvkel 3mo ago
> Vibe coding is a software development practice assisted by artificial intelligence (AI) where the software developer describes a project or task in a prompt to a large language model (LLM) which generates source code automatically. Bas
48.
by mvkel 3mo ago
In the three years I've been vibe coding, it's the only way I've found to consistently and reliably build complex apps. How do you do it?
49.
by mvkel 3mo ago
I wonder what the average battery life is on these
50.
by mvkel 3mo ago
At least it's being used. There are many examples of tech over-adoption, like building out capacity for 1M concurrent users, only to see 50.
51.
by mvkel 4mo ago
I'm not sure who this is for. It's a time trial style helmet, but isn't very useful for time trialing. And it's not a commuter helmet. And it's not a road riding helmet, as it has no breathability. Is this a defensi
52.
by mvkel 4mo ago
The glasses are a little rose-colored here. As if anything written <2022 were on stone tablets. LinkedIn slop was human generated before it was ai generated, but it was still slop. The issue with content >2022 isn't that it'
53.
by mvkel 4mo ago
Boston Dynamics has been the dancing robot demo company since its inception. I guess they held out long enough to cash out on a hype cycle. Well played by them.
54.
by mvkel 4mo ago
Are you saying that with today's hindsight, or would you be saying that at the time of its creation?
55.
by mvkel 4mo ago
Thanks for your suggestion and kind words!
56.
by mvkel 4mo ago
I love this idea; mixes creativity with practicality
57.
Show HN: Attagram, a tiny printer that gives kids a magical daily digest(attagram.com)
9 points by mvkel 4mo ago|6 comments
58.
by mvkel 4mo ago
"Revenue: 13.07b; cost of revenue: 7.5b." This includes running inference for ~1b free users. What is untrue about this?
59.
by mvkel 4mo ago
Gemini's existence disproves this
60.
by mvkel 4mo ago
My takeaway from this is that it's incredibly validating as a business model. Inference is _highly_ profitable. Of course, like any company that has ever tried to grow at breakneck pace, you run at a loss until you "win."

More ›