iiinova — where your digital self keeps evolving

Reposted by Sireesh Gururaja

Cella @cellllla.bsky.social · 21/09/2026

Please join us at Carnegie Mellon University and online on October 2nd, 2026 for the Community-Centered AI Infrastructure Summit, a statewide convening to consider the opportunities, challenges, and governance of AI infrastructure development in Pennsylvania. RSVP here: communityaisummit.org

Flyer for the Community-Centered AI Infrastructure Summit on Friday October 2, 2026 from 9-5pm at Mellon Institute Auditorium, Carnegie Mellon University, Pittsburgh, PA. Join residents, community organizations, policymakers, researchers, technologists, energy experts, and advocates from across Pennslyvania for a public convening. Organized by CMU, Data & Society, PA Climate Equity Table, and Center for Coalfield Justice, with support from the Block Center.

22616

Sireesh Gururaja @siree.sh · 09/09/2026

hindley milner tripe system

030

Reposted by Sireesh Gururaja

Joseph Fasano @josephfasano.bsky.social · 24/08/2026

Back to school. ❤️

292565549

Reposted by Sireesh Gururaja

e.w. niedermeyer @niedermeyer.online · 21/08/2026

If you want people to engage with a technology you care about, you have to find ways to frame it around providing them with some agency. You also have to be ready for people to just say "fuck you, no" and be OK with that. You have to learn to center people and not the technology itself.

1141461

Reposted by Sireesh Gururaja

P(aul) Frazee @pfrazee.com · 05/08/2026

My impression is that AI coding is on a "pick two of three" triangle: scope, ship speed, and correctness. Small tools are (speed + correctness) fast prototypes are (speed + scope) and big projects are (scope + correctness). Every project that tries to be all three degrades to "fast prototype."

2343939

Reposted by Sireesh Gururaja

New-Cleckit Dominie @ncdominie.bsky.social · 28/07/2026

This is just to say I have OCRed the text thet was in tne arch:ve and whicb yov were prabaldy seamhing for "breahtast" Fargme ms fheg mem debo:w5 sb ?mccf mmb 5c oc1b

7778522569

Sireesh Gururaja @siree.sh · 24/07/2026

It's finally out! At a time when basic research funding through the NSF and NIH in the US is under threat, it's important to understand what "basic" means in the context of more available DoD funding. For a lot of academics, the DoD's "basic" might not be be as basic as you think!

1115

Sireesh Gururaja @siree.sh · 23/07/2026

40 minute headways in Pittsburgh are fine, apparently?

Google maps directions where the transfer instruction is "Walk, then wait for up to 1455 minutes"

020

Reposted by Sireesh Gururaja

Mateusz Fafinski @calthalas.bsky.social · 18/07/2026

Remember, both the Iliad and the Odyssey were originally audiobooks.

18373481317

Sireesh Gururaja @siree.sh · 05/07/2026

Great thread from a book that has lived rent-free in my head since it argued that computers are inherently and fundamentally conservative. I didn't even know it was out of print!!

"Yes, the computer did arrive "just in time." But in time for what? In tie to save—and save very nearly intact, indeed, to entrench and stabilize—social and political structures that otherwise might have been either radically renovated or allowed to totter under the demands that were sure to be made on them. The computer, them, was used to conserve America's social and political institutions. It buttressed them and immunized them, at least temporarily, against enormous pressures for change."

0183

Sireesh Gururaja @siree.sh · 23/06/2026

Extremely hype to read this 👀

031

Reposted by Sireesh Gururaja

Brian Waismeyer @yetanotherbrian.bsky.social · 08/06/2026

good ML: * carefully designed for narrow contexts (or with awareness of context limits) * designed by ML and topical experts, with expert curated data and expert reviewed outputs * designed for specific needs and consumers, and kept up to date for both * clear training / comms on use and limits

1598

Sireesh Gururaja @siree.sh · 05/06/2026

The implicit compositionality born of English's status as a bastard language is a gift that calls for audacity, chutzpah, moxie, panache, etc

031

Reposted by Sireesh Gururaja

🌶 David Gray Widder @davidthewid.bsky.social · 03/06/2026

I am beginning new research, and would like recommendations on your favorite writing on the "strange bedfellows" phenomenon. Unlikely coalitions fighting for a common goal, and the ethical quandaries therein.

172

Sireesh Gururaja @siree.sh · 02/06/2026

Truly wonderful conversation with my lyft driver this morning, which started by discussing how he wanted to go on vacation in Florida, to see alligators in the wild. Normal, relatable, etc - until he reveals that he in fact raises alligators as pets - but only until they're eight (8) feet long!

130

Reposted by Sireesh Gururaja

rax ‘levon honkers’ king @raxkingisdead.bsky.social · 26/05/2026

‘dialectics’ is grad school for ‘there’s two of them’

3194199

Reposted by Sireesh Gururaja

Revan @revan.eurosky.social · 18/05/2026

> looking at tech sector layoffs > ask the hr manager if the layoffs are from ai or from the end of zirp > she doesn't understand > pull out illustrated diagram explaining what is ai replacement and what is zirp > she laughs and says "they’re good layoffs sir” > check the data > its zirp

657185

Sireesh Gururaja @siree.sh · 11/05/2026

As someone that came of age in a time where the linux distros I used had easy access to a *huge* variety of loyal clients (xmpp, so many web things), this paper and its framing are a lovely way to describe the thing I'd love for computing to return to ❤️

030

Sireesh Gururaja @siree.sh · 10/05/2026

It's me! I am both this type of guy and, regrettably, this guy

010

Reposted by Sireesh Gururaja

lauren @lauren.rotatingsandwiches.com · 29/04/2026

at this point anyone who sees any actual value in these tools has to be rooting for the bubble to pop so the moneyed freaks propping it up move on to quantum computing already

1767047

Sireesh Gururaja @siree.sh · 24/04/2026

Great quote from a great article, and it drives me *nuts* how big the gap is between the real affordances of AI, which can be fun and adaptive and agency affirming the political project it is inextricable from, software-brained imperialism :/

030

Reposted by Sireesh Gururaja

Efrén Pérez @efrenpolipsy.bsky.social · 22/04/2026

Me agreeing to a new project knowing full well it’s gonna add stress 🇲🇽

151760255

Reposted by Sireesh Gururaja

Stella Biderman @stellaathena.bsky.social · 22/04/2026

My hot take is that the median social system breaks under too much optimization pressure, and we should stop trying to optimize things

23923145

Sireesh Gururaja @siree.sh · 13/04/2026

This is a great thread, and reminds me of interviewing someone at one of the big labs for a study, and they made the claim that "classification was solved in 2022" That's an insane quote, until you account for the money, time, and labor that will be poured into any problem that's deemed worth it

151

Reposted by Sireesh Gururaja

GAI Debord @dylanschenker.com · 10/04/2026

They need to make it so you can mass add starter packs to a list / feed if you don't want them to fill up your main feed

111

Reposted by Sireesh Gururaja

Vicki @vickiboykis.com · 08/04/2026

friendly reminder to run docker system prune if you haven't recently

8707

Reposted by Sireesh Gururaja

Lindia Tjuatja @lindiatjuatja.bsky.social · 06/04/2026

Really excited about this work w/ my long-time collaborators at Boulder! We address limitations in existing morphosyntactic annotation systems for digitally under-resourced languages and show how *jointly* predicting morphological segmentation helps with glossing performance

063

Reposted by Sireesh Gururaja

jesse @jesseltaylor.bsky.social · 03/04/2026

if you invented public libraries today, every opinion page in the country would be arguing for means tested subsidized Amazon Prime memberships

5873711875

Reposted by Sireesh Gururaja

Sasha Costanza-Chock @schock.cc · 02/04/2026

It's time to seize the means of prediction

14013

Reposted by Sireesh Gururaja

Stella Biderman @stellaathena.bsky.social · 29/03/2026

You have a moral imperative to refuse to work with these people or develop models for these purposes.

193

Sireesh Gururaja @siree.sh · 24/03/2026

*CL folks, a recent history question: was the move to OpenReview as the review platform for *CL conferences related to the move to ARR? Or did they just happen simultaneously? Asking for a discussion with @shaily99.bsky.social

010

Sireesh Gururaja @siree.sh · 23/03/2026

I'm really not a fan of the way some journals in the physical sciences state results upfront, and bury methods much later in the paper, or even in supplementary materials. Maybe it's my ML bias, but I don't trust that your methods are prima facie reasonable!! Show me what you did before the results!

021

Reposted by Sireesh Gururaja

maybe: leif weatherby @leifw.bsky.social · 21/03/2026

Appointment reading

094

Reposted by Sireesh Gururaja

William B. Fuckley @opinionhaver.bsky.social · 21/03/2026

I think my AI beliefs are quickly becoming ‘this has enormous positive potential if implemented responsibly. it will not be implemented responsibly and like maybe five companies are even trying.’

2749461

Sireesh Gururaja @siree.sh · 26/02/2026

This is a great thread, and makes me think about how the affordances of LLMs (largely text-only chat) preclude the kind of epistemic vigilance we were able to develop around tech like search. I think we can build LLM-based tech that's good here, but chat ain't it

081

Reposted by Sireesh Gururaja

Colin @colin-fraser.net · 20/02/2026

Rename LASSO and Ridge to kiki and bouba

191

Sireesh Gururaja @siree.sh · 13/02/2026

This does feel like situation where open models are held back by the state of benchmarks. For an open model, there isn't as much of a theory of use outside of benchmarks. For closed models, the companies putting them out can build for business cases and their associated internal benchmarks

110

Reposted by Sireesh Gururaja

Jake Grumbach @jakemgrumbach.bsky.social · 11/02/2026

Most tech workers, even at Palantir (probably), are pro-social compared to the tech owner and VC classes. They need unions so their voices matter for company governance. www.wired.com/story/palant...

wired.com

Palantir CEO Alex Karp Recorded a Video About ICE for His Employees

In a video shared with Palantir employees, Alex Karp did not explain how ICE is utilizing the company’s products. Instead, workers were told they can sign NDAs if they want detailed information.

112822

Sireesh Gururaja @siree.sh · 10/02/2026

Vivid mental image of cellino and barnes being like penn and teller, where one is always completely silent and if you advance a surprising legal theory your consultation is free

010

Sireesh Gururaja @siree.sh · 10/02/2026

Firewatch is one of the most lovely games I've played. I've taken so much joy from playing it with multiple friends and seeing how different a game it is for them. I didn't know games could tell stories like this before it (It also played a small part in my fiancée and me getting together!)

072

Sireesh Gururaja @siree.sh · 03/02/2026

This is a real banger of a paper. The example of a model being weirdly focused on jasmine (lol) makes me increasingly think that single-point-of-access models don't really consider who their audience is. Jasmine is a super legible cultural marker for people outside, but is so, _so_ generic.

2124

Reposted by Sireesh Gururaja

naitian @naitian.org · 29/01/2026

“Position paper” is just a label to make some kinds of interdisciplinary theoretical work fit into the CS publishing schema.

2113

Sireesh Gururaja @siree.sh · 26/01/2026

extremely on brand, love this town

050

Reposted by Sireesh Gururaja

Shannon Vallor @shannonvallor.bsky.social · 22/01/2026

Best use of this meme i’ve ever seen

14417

Sireesh Gururaja @siree.sh · 20/01/2026

Another trip to Bangalore, another time that Google Maps (but no other Google service) turns itself Indian accented and maybe can't be turned back? 🤔

071