Aditya Sengupta
@adiastra.bsky.social
560 followers 610 following 160 posts
Astronomy PhD student at UCSC
Posts Media Videos Starter Packs
adiastra.bsky.social
ignoring float warnings is a time honored tradition (also eventually getting them to do what you want using haphazard \vspace-s)
adiastra.bsky.social
as someone not really in this field but w experience with Kalman filters, do you know of a reference that talks about how the EnKF is used in numerical weather prediction?
adiastra.bsky.social
(eric adams voice) new york is the springfield of polynesia
adiastra.bsky.social
isn't making statistical models exactly the kind of programming LLMs are meant to be good at? Doesn't sound at all unconventional
adiastra.bsky.social
it's sort of a throwaway line but "analyzing our bloodwork" makes me think there's some conflation of genAI with other kinds of machine learning going on here too. Implicitly taking the credit for types of ML that have worked for years is a big part of the genAI marketing strategy!
adiastra.bsky.social
the whole bit from Echoes by Pink Floyd that's just radar noises
Reposted by Aditya Sengupta
astroshashank.bsky.social
New paper by me and @benjaminpope.bsky.social! How well can we map a star using optical interferometry?

This is just a submitted preprint at the moment--comments and questions welcome! (1/N)

arxiv.org/abs/2509.25433
adiastra.bsky.social
there's a sign not far from me on campus that says "I am more than my productivity" and every time I see it I go "ok but they mean everyone but me though right"
Reposted by Aditya Sengupta
retr0.id
have you seen the new python package? it's on pypi. you can literally install it with pip. set up a venv. grab it with poetry. you just use rye. run it with pipx. you can install it with uv. install uv right now. go to uv. dive into uv. you can uv it. it's on uv. uv has it for you. uv has it for you
adiastra.bsky.social
oh NO you did a B side before 3A? you're so powerful lol
adiastra.bsky.social
yeah, top of 3A I think. On the little loading screen postcard
adiastra.bsky.social
like they say, be proud of your death count!
adiastra.bsky.social
curious here, do you know of a good review of theoretical neuro for someone who knows physics/math?
adiastra.bsky.social
this is all cool as hell i want this art for my wall
adiastra.bsky.social
you're actually so right i want Texture in food
adiastra.bsky.social
a couple months ago i started hearing goats out my window at 11pm (in Santa Cruz), thought i was going insane, and spent 40 minutes walking around trying to triangulate them, turns out it was this
Reposted by Aditya Sengupta
arula-ratnakar.bsky.social
I actually kind of want to hear from people who dislike mathfiction.

What stops you from engaging with a story with math in it? What would draw you to trying one out?

#Writersky #Writing #SFF
arula-ratnakar.bsky.social
But it will have math— that’s not going away in my stories ever. I personally think mathematical beauty is worth studying in stories, just like love, or friendship, or ambition, or cosmic awe, or death, or literally any other source of inspiration for literature.
adiastra.bsky.social
sooo true! I think it's so hard to explain why people find math beautiful, and I never really thought of literature as a way of getting that across but would love to see more of that
adiastra.bsky.social
Really wondering what AI is gonna think cricket is about
Chat summary (of a YouTube live chat)
Auto-generated experiment, quality may vary 
Viewers are excited about a wicket and praising the bowler, referring to him as "DSP". There's discussion about a player named Crawley being "arrested".
Reposted by Aditya Sengupta
adiastra.bsky.social
garmin has sleep quality awards?? I'm nowhere near this clearly lmao
adiastra.bsky.social
I don't even know what the supposed mathematician logic would be here?? There's something deeper than just not understanding math here - they think they do understand something others don't when it's exactly the opposite
adiastra.bsky.social
analytic and algebraic topology of locally euclidean metrization of infinitely differentiable riemannian manifold
adiastra.bsky.social
I don't necessarily think it has to be the same thing but it does have to be batch cooked - when I meal prep I do 3-4 different things and decide how to combine them into actual dishes based on vibes