aadarsh

@uchicago @mitqpg | prev @skyroota

[aadarsh @aadarwal]

@itamarwe @docmilanfar @CSProfKGD oh ok

[aadarsh @aadarwal]

@itamarwe @docmilanfar @CSProfKGD i think you mean theta here is what should be the same as the zoom ? that would make sense but they aren’t doing that it’s a fixed zoom?
ig the theory is applicable but it’s v diff example

[aadarsh @aadarwal]

@itamarwe @docmilanfar @CSProfKGD but this doesn’t make sense the helicopter isn’t moving like the rig ?

[Skyroot Aerospace @SkyrootA]

Aagaman means "Arrival." ✨

Proud to unveil the #MissionAagaman patch — the arrival of a new Indian rocket on the world stage, and a new chapter for all of us at Skyroot.
The patch honors India's space journey, and the era of private orbital launches it opens.
Vikram-1, Test Flight-1.
Onwards & Upwards. 🚀

#Vikram1 #OpeningSpaceForAll

[QC @QiaochuYuan]

sam kriss on pangram (i have also been, frankly, baffled at the anti-pangram backlash)

[aadarsh @aadarwal]

@jaibellare @hackerfabindia sick af

[Sparsh Agarwal @sparshselim]

The Age of AI demands planetary-scale Carbon Dioxide removal.

@AltCarbonIndia is using volcanic rock dust to geochemically pull carbon out of the atmosphere — and we just proved it works at scale. The world's largest issuance of carbon credits through Enhanced Rock Weathering.

~10,000 tonnes of CO₂ removed. Enough to offset a small AI data centre.

India has a history of scientific breakthroughs that stun the world — across medicine, space exploration, energy, & financial inclusion.🇮🇳

Climate Change is the most significant existential threat to our species. It demands Himalayan Ambitions.

We're moving mountains to make that happen. Literally.

[Peter Schmidt-Nielsen @ptrschmdtnlsn]

I'm very curious if they 1) have some new mechanism to radically reduce the rate of overconfient "aha, I solve it" false positives, 2) did some automated searching for the diamond among the slop, or 3) did tons of human reveiw.

[aadarsh @aadarwal]

@Tyler_A_Harper thanks!

[Gabriel @gbrl_dick]

one of the reasons that I find myself disinclined to read obviously ai-generated writing is that, regardless of how it's phrased (and the models tend towards absolutism), it tends to be a restatement of the consensus view that i already know, or have easy access to.

i don't know exactly how this will translate to the world roon is imagining - one where models with much greater capabilities are given wide latitude over strategy and direction. but i suspect that the general trend will hold, and that, even as the models get super-human along a bunch of capabilities, they will retain a kind of generality in their thinking that will make them much better advisors than actors.

[aadarsh @aadarwal]

@theo icloud desktop :(

[Hugging Models @HuggingModels]

A massive 30 billion parameter MoE model just landed on Hugging Face. It's a compressed, quantized version of Sarvam AI's open source model. This is a big deal for Indian language AI and efficient inference.

[ThePrimeagen @ThePrimeagen]

@ZachFergus I think you need the first answer the question of what is the point of working for a problem before you can answer how long

[aadarsh @aadarwal]

@ethayarajh @Diyi_Yang @ylecun sick

[aadarsh @aadarwal]

@adadithya use wispr flow its the best!

[aadarsh @aadarwal]

@adadithya “abundance”:)

[Alt Carbon @AltCarbonIndia]

We get a lot of design questions about how we build our microsites.

@geer_raah wrote a post explaining our process.

⋆ Reconstructed faces from faded 19th century photographs.

⋆ Built a 3D Jorasanko Thakur Bari — the Tagore ancestral home — as the spine of the site.

⋆ Reimagined Raphael's School of Athens, set it in Bengal, and populated it with 13 key figures.

⋆ Built motion flowers for Begum Rokeya's 1905 feminist utopia.

⋆ Sat with Chittoprasad's famine sketches until the pain of it dawned on us all.

This is the design story behind our Bengal Renaissance Project microsite — and an honest account of everything that broke along the way.