side_up_down

4 ms·

1.
by side_up_down 8mo ago
Brave has an AI-specific API
2.
by side_up_down 8mo ago
I'd take "running at home" to mean running on reasonably available consumer hardware, which your setup is not. You can obviously build custom, but who's actually going to do that? OP's point is valid
3.
by side_up_down 10mo ago
How does this incorporate case law?
4.
by side_up_down 10mo ago
There's a great non-AI point in this article - Seattle has great engineers. In pursuing startups, Seattle engineers are relatively unambitious compared to the Bay Area. By that I mean there's less "shooting for unicorns"
5.
by side_up_down 1y ago
There are thousands of Meta employees on HN and none answered this post. Why not? Look at the kinds of discussion that happened in their absence. I worked at Meta for a few years. Facsimiles aside, it's no different than any other corp
6.
by side_up_down 2y ago
Meta Reality Labs - Conversational AI | https://www.metacareers.com/jobs/2780884748740452/ | Research Engineer | Machine Learning Engineer | Redmond, WA; Menlo Park, CA | Full-time | ONSITE I'm hiring Researc
7.
by side_up_down 3y ago
You can be sure Google's lawyers vetted the training to ensure it was lawful instruction.
8.
by side_up_down 4y ago
It sounds like there's some other cultural problems at your company :)
9.
by side_up_down 5y ago
Nice framing. I agree with that interpretation. I'd argue communication is always an approximation to an ideal, anyways. The original version is going to be incomplete. How adequately can English (or your language of choice) represent
10.
by side_up_down 5y ago
I'd add another sentence to that: "Understand that even with the best intentions you'll likely still get it wrong." Objective truth is incredibly unwieldly. It's almost impossible to get right - just considering imp
11.
by side_up_down 6y ago
What?
12.
by side_up_down 7y ago
Very pithy comment. Breakthroughs are only breakthroughs in hindsight. False positives and negatives abound in breakthrough declarations.
13.
by side_up_down 8y ago
Empericially, CNNs generalize better on image recognition tasks than hand built features. This comment doesn't make much sense and is needlessly obtuse in the face of progress, tbh.
14.
by side_up_down 8y ago
Thats a great answer but decidedly off topic. He asked a very specific question and you gave a completely different answer. HN, this shouldn't be voted up here. Our biases are showing by doing so.
15.
by side_up_down 9y ago
> I wish hacker news got excited about the filling the pipe post and less excited about this thread. That's disingenuous considering the number of votes and comments each thread have received. HN IS more excited by the pipe post as
16.
by side_up_down 9y ago
Big advance? AI than can learn conditional logic. DL tops out as mapping vector spaces to less complicated vectors. Incredibly powerful, but incredibly limited in what problems it can emulate.