Andrew Bean

2.9K posts

Andrew Bean banner
Andrew Bean

Andrew Bean

@AndrewBean

ML Engineer @SpaceX formerly @xAI formerly @X formerly @Twitter | PhD from @ECEILLINOIS

Boston, MA Katılım Haziran 2018
164 Takip Edilen907 Takipçiler
Crémieux
Crémieux@cremieuxrecueil·
It's interesting how there seem to be two perspectives on the 'big goal' for intelligence research. In America, there's more focus on enhancement. In Britain, there's more focus on preventing cognitive decline.
English
6
1
149
8.9K
i/o
i/o@avidseries·
He's the kind of Christian who prefers the tribal, capricious, vengeful, and genocidal God of the Old Testament to the forgiving and merciful God of the New Testament.
English
32
2
114
5K
Matthew Zeitlin
Matthew Zeitlin@MattZeitlin·
we have one of these in nevada, it's more or less universally considered to have been a complete boondoggle because normal solar and batteries have gotten so much cheaper (also a lot of environmentalists hate it)
Matthew Zeitlin tweet media
Seth Harp@sethharpesq

To the immiserated proles of the Epstein Empire, this doesn't even look real. In the United States, "advanced" technology is a bumbling Waymo that won't get out of the way, or the Unit 8200 app that scans your face at the airport.

English
42
48
1.5K
132.3K
10x Engineer
10x Engineer@10x_er·
What to do in Vegas if you don’t drink. Should I even go?
English
33
1
30
3.6K
Andrew Bean
Andrew Bean@AndrewBean·
@blind_via Or perhaps one main and several peripheral cameras stitched, with responsive digital view angle changes and best effort to physically point the main camera in the view direction
English
0
0
2
26
Andrew Bean
Andrew Bean@AndrewBean·
@blind_via It should just have 360 camera coverage and not need to actually turn
English
2
0
8
166
Bob
Bob@root_of_trust·
@moultano The yellow line is PCA
English
1
0
1
165
Ryan Moulton
Ryan Moulton@moultano·
There's a pattern where people try to calibrate one measure to another using regression, which isn't "wrong" in a predictive sense, but doesn't behave the way people expect unit conversion to work. Every subsequent conversion narrows the range, and the conversion isn't invertible
Ryan Moulton tweet media
English
14
5
104
9.3K
kache
kache@yacineMTB·
you still need to read and understand code
English
220
264
3.6K
135.2K
Jack
Jack@jackunheard·
REPORTER: "What makes you trust Iran?!" TRUMP: "I don't trust ANYBODY. I don't trust YOU." Never gets old 🤣
English
230
2.8K
25.8K
307.2K
Andrew Bean
Andrew Bean@AndrewBean·
@blobdell I wrote phonon simulation in Fortran long ago. That was my only use of it ever
English
0
0
0
10
Andrew Bean
Andrew Bean@AndrewBean·
@i2cjak @GreeneElizabeth I think there's more to it than that, eg common resistor values work well for inverting or non-inverting amplifiers for typical/round gain values
English
0
0
0
31
i2cjak
i2cjak@i2cjak·
valid resistor values: 4.7k 5.1k 10k 100k 1M mental illnesses: 1.5k ??? 3.3k 6.8k
Español
55
21
561
15.7K
Andrew Bean
Andrew Bean@AndrewBean·
@jsuarez @KyleMorgenstein I don't understand how people fall into this moral relativism paralysis. I assure you our enemies do not have this problem.
English
0
0
0
93
Joseph Suarez 🐡
Joseph Suarez 🐡@jsuarez·
@KyleMorgenstein We're not the good guys. There probably are no good guys. But there are absolutely bad guys. It is that simple.
English
2
0
13
327
Kyle🤖🚀🦭
Kyle🤖🚀🦭@KyleMorgenstein·
I was told this argument many many times when I put my foot down about never taking military funding for my PhD in *aerospace engineering* despite having to TA/take jobs/extend my degree to fund it. I’m glad I never took it.
Jeremy Noronha@JeremyNoronha

@imperialauditor If you say no, someone with less values or kindness than you ends up with the power and money. The world is better off it being you cause you have a higher chance of making something good come from it.

English
4
2
90
12.6K