Andy Timm
@andytimm.bsky.social
Data scientist @ Grow Progress, testing Dem ads. Usually thinking about methods for survey weighting, efficient causal inference, or scalable Bayes. https://andytimm.github.io/
Just here to thank you for the public service of posting this.
Was wondering about some of these checks!
Was wondering about some of these checks!
December 13, 2025 at 3:56 PM
Just here to thank you for the public service of posting this.
Was wondering about some of these checks!
Was wondering about some of these checks!
Don’t think these are new methods?
I’m pretty pessimistic about these directions both because:
1. These are really unoptimized bot prompts/setups!
2. The labs (rightly) understand addressing various injection-like attacks as key to making browser use more valuable: www.anthropic.com/research/pro...
I’m pretty pessimistic about these directions both because:
1. These are really unoptimized bot prompts/setups!
2. The labs (rightly) understand addressing various injection-like attacks as key to making browser use more valuable: www.anthropic.com/research/pro...
Mitigating the risk of prompt injections in browser use
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
www.anthropic.com
December 8, 2025 at 1:32 PM
Don’t think these are new methods?
I’m pretty pessimistic about these directions both because:
1. These are really unoptimized bot prompts/setups!
2. The labs (rightly) understand addressing various injection-like attacks as key to making browser use more valuable: www.anthropic.com/research/pro...
I’m pretty pessimistic about these directions both because:
1. These are really unoptimized bot prompts/setups!
2. The labs (rightly) understand addressing various injection-like attacks as key to making browser use more valuable: www.anthropic.com/research/pro...
I thought it was the probability of astronomical suffering (this joke is really for like 2 people so will just link this: en.wikipedia.org/wiki/Risk_of...
Risk of astronomical suffering - Wikipedia
en.wikipedia.org
November 25, 2025 at 8:46 PM
I thought it was the probability of astronomical suffering (this joke is really for like 2 people so will just link this: en.wikipedia.org/wiki/Risk_of...
Ah yeah that’s def preferable if you don’t have a reason to build yourself.
November 19, 2025 at 9:56 PM
Ah yeah that’s def preferable if you don’t have a reason to build yourself.
As a practical solve, are you setting TORCH_CUDA_ARCH_LIST for your particular GPU before the install? You can google the right version for your GPU.
Can say more about what causes this if helpful, I got a bit nerdsniped a while back by how heavy templating makes the kernels so big.
Can say more about what causes this if helpful, I got a bit nerdsniped a while back by how heavy templating makes the kernels so big.
November 19, 2025 at 9:26 PM
As a practical solve, are you setting TORCH_CUDA_ARCH_LIST for your particular GPU before the install? You can google the right version for your GPU.
Can say more about what causes this if helpful, I got a bit nerdsniped a while back by how heavy templating makes the kernels so big.
Can say more about what causes this if helpful, I got a bit nerdsniped a while back by how heavy templating makes the kernels so big.
@seanjwestwood.bsky.social maybe can share?
November 18, 2025 at 1:18 PM
@seanjwestwood.bsky.social maybe can share?
Been a minute since I lived there, but a few ideas— Fairstate’s probably my favorite, mix of great IPAs and some odder (but well done) experimental-ish beers.
Also great are Lake Monster (a lot of really solid, more sessionable stuff), and Lynnlake (great at a lot of styles that I love for fall).
Also great are Lake Monster (a lot of really solid, more sessionable stuff), and Lynnlake (great at a lot of styles that I love for fall).
September 9, 2025 at 1:02 AM
Been a minute since I lived there, but a few ideas— Fairstate’s probably my favorite, mix of great IPAs and some odder (but well done) experimental-ish beers.
Also great are Lake Monster (a lot of really solid, more sessionable stuff), and Lynnlake (great at a lot of styles that I love for fall).
Also great are Lake Monster (a lot of really solid, more sessionable stuff), and Lynnlake (great at a lot of styles that I love for fall).
You don’t need to compile all your packages on Linux anymore :)
Check out pak/p3m, which has pre-compiled binaries: r-in-production.org/packages.html
Check out pak/p3m, which has pre-compiled binaries: r-in-production.org/packages.html
6 Package installation – R in Production
r-in-production.org
July 23, 2025 at 11:42 AM
You don’t need to compile all your packages on Linux anymore :)
Check out pak/p3m, which has pre-compiled binaries: r-in-production.org/packages.html
Check out pak/p3m, which has pre-compiled binaries: r-in-production.org/packages.html
Kevin Munger has both some great descriptive work and meta scientific argument articulating the value of descriptive work for fast moving topics like social media platforms in his book The YouTube Apparatus!
See also his temporal validity paper for an articulation of the metascientific point.
See also his temporal validity paper for an articulation of the metascientific point.
May 29, 2025 at 3:35 PM
Kevin Munger has both some great descriptive work and meta scientific argument articulating the value of descriptive work for fast moving topics like social media platforms in his book The YouTube Apparatus!
See also his temporal validity paper for an articulation of the metascientific point.
See also his temporal validity paper for an articulation of the metascientific point.
What you’re looking for*
May 15, 2025 at 10:35 PM
What you’re looking for*
Omg, need to get my hands on one of these tomorrow or Friday. Where were these?
May 14, 2025 at 11:53 PM
Omg, need to get my hands on one of these tomorrow or Friday. Where were these?
👋 First time going, I’m excited!
May 13, 2025 at 11:19 AM
👋 First time going, I’m excited!
Reposted by Andy Timm
Could. Not. Resist.
May 9, 2025 at 10:49 AM
Could. Not. Resist.
Definitely read this as Y-MRP.
Yung Multilevel Regression and Postratification seems intriguing 🤔
Yung Multilevel Regression and Postratification seems intriguing 🤔
April 14, 2025 at 12:05 PM
Definitely read this as Y-MRP.
Yung Multilevel Regression and Postratification seems intriguing 🤔
Yung Multilevel Regression and Postratification seems intriguing 🤔
Oh god, am I going to do this as a side project for the bit?
Footnotes of footenotes? call them… toes
Footnotes of footenotes? call them… toes
February 27, 2025 at 1:38 AM
Oh god, am I going to do this as a side project for the bit?
Footnotes of footenotes? call them… toes
Footnotes of footenotes? call them… toes
I need footnotes within footnotes, that’ll make everything much more hinged
February 27, 2025 at 12:28 AM
I need footnotes within footnotes, that’ll make everything much more hinged
Pivoting slightly, you said “God, what have I done?”, so you’re now a pink pony girl and you dance at the club.
youtu.be/8lmK_dMWY8o?...
youtu.be/8lmK_dMWY8o?...
Watch CHAPPELL ROAN perform "PINK PONY CLUB" live at the 2025 GRAMMYs
YouTube video by Recording Academy / GRAMMYs
youtu.be
February 19, 2025 at 12:22 PM
Pivoting slightly, you said “God, what have I done?”, so you’re now a pink pony girl and you dance at the club.
youtu.be/8lmK_dMWY8o?...
youtu.be/8lmK_dMWY8o?...
The example that motivated this post was “making actually good simulation studies”, which I was lucky to get in grad school, but lots of folks didn’t it seems.
February 17, 2025 at 3:24 PM
The example that motivated this post was “making actually good simulation studies”, which I was lucky to get in grad school, but lots of folks didn’t it seems.
I appreciate the correction here- will say it’s very hard to be like “yep that’s fake news” when the real news is like “Musk pushes refugee status for white Afrikaners”.
February 13, 2025 at 5:16 PM
I appreciate the correction here- will say it’s very hard to be like “yep that’s fake news” when the real news is like “Musk pushes refugee status for white Afrikaners”.
Ordered two books as a result of this thread :)
February 13, 2025 at 12:51 AM
Ordered two books as a result of this thread :)