mgdev

7 ms·

91.
by mgdev 2y ago
"Save while you're earning decent cash" seems pretty actionable to me.
92.
by mgdev 2y ago
I think "aligning super intelligence" is a nothingburger of a goal, for exactly that reason. It's not a problem that's unique to OpenAI. The reason you can't "align" AI is because we, as humans on the plan
93.
by mgdev 2y ago
I like this take best. :) Most folks who are part of the startup, don't care to stick around for the scale. Different phases of the company tend to attract different types of people.
94.
by mgdev 2y ago
I hadn't until now, but it's the other failure mode I mentioned in another fork [1] of this thread: > A small, central R&D team may work with management to set the bar, but they can't be responsible for mitigating the
95.
by mgdev 2y ago
Oh?
96.
by mgdev 2y ago
> The purpose of the risk team at OpenAI was to prevent the destruction of humanity. Yeah, the problem (in this outsider's opinion) is that that charter is so ill-defined that it's practically useless, which in turn means that
97.
by mgdev 2y ago
> Imagine if vehicle manufacturers[2] split their design and R&D teams into a "make the thing go" team and a "don't kill the passengers" team. Rocket caskets. Can't kill someone who is already dead!
98.
by mgdev 2y ago
Agree that you need checks and balances, but there are better and worse systems. As someone who has scaled orgs from tens to thousands of engineers, I can tell you: you need value teams to own their own risk. A small, central R&D team m
99.
by mgdev 2y ago
Yes, it's valuable to have a small research team who focuses on R&D outside the production loop. But when you give them a larger remit, and structure teams with some owning "value" and and others essentially owning "
100.
by mgdev 2y ago
This will be a boon for Oracle.
101.
by mgdev 2y ago
Nothing new. Amazon's (now discontinued) browser extension [1] was, among other things, used to get competitive price information w/o being hit by crawling restrictions. [1]: https://www.tomsguide.com/news/amaz
102.
by mgdev 2y ago
Reading the actual case document, it's hard to chalk this up to anything but gross incompetence. Dude gets off the phone after learning the information, and immediately calls the brokerage to write puts that expire 11 days later (strik
103.
by mgdev 2y ago
I genuinely can't tell if George Hotz is genius or, mid and totally full of himself.
104.
by mgdev 3y ago
Time to gather the dragon balls.
105.
by mgdev 3y ago
Turn on FileVault. Don't give anyone your password. Pretty fathomable.
106.
Some Lessons Learned Building LLM-Powered Agents(mg.dev)
1 points by mgdev 3y ago|0 comments
107.
by mgdev 3y ago
They're focusing on what matters.
108.
by mgdev 3y ago
Rustler is great. Though this gets me thinking about how you can maintain as many Elixir invariants and conventions as possible, even while escaping them under the covers. Being able to call FeGraph.set/2 and have db actually be mutate
109.
by mgdev 3y ago
Talking to Goody-2 reminds me of working at Meta (during the ZIRP era).
110.
by mgdev 3y ago
Agree. They failed to capitalize on their existing brand awareness. All it would have taken was a coordinated, concentrated, 3-4 week GTM motion leading up to product availability.
111.
by mgdev 3y ago
Building deck = programming Playing = executing
112.
by mgdev 3y ago
The sample code plus walkthrough sounds like "things people can run on their computers". :)
113.
by mgdev 3y ago
Let's call it the "messy middle" and see if we're on the other side of it. :)
114.
by mgdev 3y ago
I would. MS has had a storied past when it comes to frontend tech, but rather than diverging and dying, what we seem to see now is convergence of a coherent technical vision (and some pruning as a result). They look the same on the surface,
115.
by mgdev 3y ago
The problem is the system on the other end usually has timing/causality expectations (e.g., in the case of reentrant calls), and that system is not usually part of your system (e.g., they wouldn't include whatever monotonically in
116.
by mgdev 3y ago
You've just pushed the problem down to the system that checks idempotence keys.
117.
by mgdev 3y ago
Apple could afford to put out the best repair tools in the industry, doing two things: 1/ deposition competitors with inferior repairability v performance scores, ultimately leading to increased market share among the most discerning c
118.
by mgdev 3y ago
As with ever other major innovation, it will both destroy and create jobs. Whether AI creates more good than harm will be hard to actually evaluate. Like every innovation, there will be general disagreement on what constitutes "good&qu
119.
by mgdev 3y ago
GraalVM is a hugely under-appreciated piece of technology. A lot of my initial interest came from being able to blend different languages into a single runtime via the polyglot APIs, but the combination of performance, strong sandboxing sup
120.
by mgdev 3y ago
Totally. I support a team of 40+ and even I have a hard time proposing this. Very hard to go against the herd. There's so much progress happening in existing Python-powered ML community that, unless there are very good interop tools an

More ›