Adam Visokay
@avisokay.bsky.social
phd candidate at uw seattle & max planck institute in rostock, germany | ai/ml, science of science, economic sociology | uva and syracuse alum | i also run. https://avisokay.github.io/ https://www.strava.com/athletes/9804160
I’m collaborating on a comparative historical piece that builds on breimans two cultures to trace the ascendancy of what we are calling the computational data regime. Should have a preprint to share soon 👀
January 7, 2026 at 3:52 PM
I’m collaborating on a comparative historical piece that builds on breimans two cultures to trace the ascendancy of what we are calling the computational data regime. Should have a preprint to share soon 👀
I was uninitiated until age 28 but now there’s no going back
January 4, 2026 at 1:10 AM
I was uninitiated until age 28 but now there’s no going back
I found this useful re custom claude agents. www.youtube.com/watch?v=RivV...
December 31, 2025 at 3:03 PM
I found this useful re custom claude agents. www.youtube.com/watch?v=RivV...
Many analysts problem? Never heard of them
December 30, 2025 at 2:33 AM
Many analysts problem? Never heard of them
Reposted by Adam Visokay
Except the statistics are bad?
Also they open by citing Galton on the heredity of genius, which sort of gives away the eugenic thinking behind it.
Also they open by citing Galton on the heredity of genius, which sort of gives away the eugenic thinking behind it.
This smells distinctly like collider bias and/or selection bias and/or regression to the mean... You simply can't select teen prodigies, and world class athletes rom databases, and go run regressions without serious consideration of the selection process!
"Most top achievers (Nobel laureates and world-class musicians, athletes, chess players) demonstrated lower performance than many peers during their early years. Across the highest adult performance, peak performance is negatively correlated with early performance" www.science.org/doi/10.1126/...
December 20, 2025 at 9:45 PM
Except the statistics are bad?
Also they open by citing Galton on the heredity of genius, which sort of gives away the eugenic thinking behind it.
Also they open by citing Galton on the heredity of genius, which sort of gives away the eugenic thinking behind it.
Hey that’s pretty neat
December 17, 2025 at 6:53 AM
Hey that’s pretty neat
Growing up in America I believed that Germans were stereotypically on time. But now after several visits I have learned that the rail system is actually quite inconsistent. (Hi from the max planck institute👋)
December 12, 2025 at 10:36 AM
Growing up in America I believed that Germans were stereotypically on time. But now after several visits I have learned that the rail system is actually quite inconsistent. (Hi from the max planck institute👋)
Reposted by Adam Visokay
Isn't an important part of the work of organizing building that imagination -- helping people see that it doesn't have to be this way? (And then also finding and sharing resources at a local level and...)
December 9, 2025 at 2:22 PM
Isn't an important part of the work of organizing building that imagination -- helping people see that it doesn't have to be this way? (And then also finding and sharing resources at a local level and...)
Have you had a chance to read this yet? www.pnas.org/doi/full/10....
The potential existential threat of large language models to online survey research | PNAS
The advancement of large language models poses a severe, potentially existential threat
to online survey research, a fundamental tool for data coll...
www.pnas.org
December 1, 2025 at 8:35 PM
Have you had a chance to read this yet? www.pnas.org/doi/full/10....
I’m new to using agents, but have a research task where i think they could be useful. given a big dataset, i want to use agents to simulate a bunch of treatment combinations - more than you could do irl. And then characterize what you can and can’t learn using agents for scientific discovery
December 1, 2025 at 7:44 PM
I’m new to using agents, but have a research task where i think they could be useful. given a big dataset, i want to use agents to simulate a bunch of treatment combinations - more than you could do irl. And then characterize what you can and can’t learn using agents for scientific discovery