dbish

21 ms·

271.
by dbish 3y ago
Models != apps, they’re closer to the backend for apps or a core library, so your customers aren’t “everyone with a phone/computer” like an app, it’s the much smaller, but potentially high impact, “everyone who creates apps”. We’re not
272.
by dbish 3y ago
This isn’t sustainable though. Compute is expensive, maybe they’ll end up with paid plans for anyone trying to use a space
273.
by dbish 3y ago
Measuring words and loc is not a great way, imho, to share what you’re doing. In fact, I’d love a much shorter set of documentation now to under and this better. Long docs will probably make it less likely to read.
274.
by dbish 3y ago
GPt-3 is very expensive if you use it frequently compared to just running in a desktop machine you already have. Of course, if you’re buying new hardware just to run a model for yourself locally, that’s a different cost analysis, but for me
275.
by dbish 3y ago
Azure is hosting and operating the service themselves rather then for OpenAI, with all the security requirements that come with that. I assume this comes with different data and access restrictions as well and ability to run in secured inst
276.
by dbish 3y ago
This is pushed from the smallest startup upwards as a way to make sure you’re building something people want. There are pros and cons of course, but I can understand why companies do this and why software folks think they need to. It’s rela
277.
by dbish 3y ago
This is why I mostly don’t worry about what they’re up to as a competitor in a particular product space. It’s rare that their advertised features match (with good quality) their released features.
278.
by dbish 3y ago
I don't deny it's awesome, but this is Varda's lab (pretty well funded VC startup). They're inspirational, but quite far from a lab :)
279.
by dbish 3y ago
Garage? That’s an industrial lab
280.
by dbish 3y ago
Arc is neat but since it’s built on chromium you’re still beholden to their decisions.
281.
by dbish 3y ago
Nice. This is legitimately a new browser. Hadn’t heard of them. Will check it out.
282.
by dbish 3y ago
I’d bet someone else owned it, had it pointed at chatgpt hoping they’d buy it, and just made a ton selling to Elon and crew for the vanity domain (which is pretty funny since x.ai is actually shorter).
283.
by dbish 3y ago
Brave is built on chromium. Cool concept, but not a new browser
284.
by dbish 3y ago
The problem is there are almost 0 alternatives beyond Firefox so if you don’t like their UX you’re stuck. All the “new” browsers are just chromium skins (edge, arc, etc.) Are there any real competitors building a new browser?
285.
by dbish 3y ago
Now let’s see how much that cost
286.
by dbish 3y ago
Folks who don’t understand market research or love the chaos? I for one thing most human based market research is low quality anyway, so why not just make something up.
287.
by dbish 3y ago
Not any of the AI features (ask/query/etc.).
288.
by dbish 3y ago
You would lose all of the natural language q&a (ask X) which imho is the useful piece.
289.
by dbish 3y ago
Probably not. These vendors have massive supply lines that are designed and tuned to their needs.
290.
by dbish 3y ago
OpenAI is pretty likely working on their own (see Kaparthy's "Building a kind of JARVIS @ OреոΑӏ"), and Microsoft of course is doing an integration or reinterpretation of Cortana with OpenAI's LLMS (since they are incapa
291.
by dbish 3y ago
You can use that, but there are pretty well understood terms in finance already (like retail investors), and probably best to use the standard if you want to be well understood.
292.
by dbish 3y ago
That’s why you base it on SAT and the like, and drop the other stuff. Studying for the SAT doesn’t have a huge impact
293.
by dbish 3y ago
They should get free tuition for their kids but that shouldn’t guarantee them a spot and take away a spot from someone more aligned to the school’s selectivity.
294.
by dbish 3y ago
Not the Iveys. They have very large scholarships if your family can’t afford it and you get in.
295.
by dbish 3y ago
Yes, that should not be allowed. The best should get in, not a relative.
296.
by dbish 3y ago
Lawyers are certainly an old profession stuck in old ways. I agree
297.
by dbish 3y ago
For students, sure, but when being hired (or funded) they all have similar “elite” tier intelligence/potential, nothing about ruling class
298.
by dbish 3y ago
See also the massive grade inflation at these schools
299.
by dbish 3y ago
Is MIT still worth the price of admission? I would certainly say so and they don't do legacy, and I would trust the Princeton/Harvard line on a resume a lot more if it was guaranteed they didn't get in because of non-academic
300.
by dbish 3y ago
I'm seeing this a lot. It's a great example of how the older non-FAANG large companies are going to just keep making it harder to actually be effective. Good oppurtunity for startups who are competing with them though as the AI to

More ›