ishan

7.9K posts

ishan banner
ishan

ishan

@usrbinishan

20. gotta make it feel home (it's working). ♥️ IL-BLR-ISK-HYD-BLR-DXB-WA-IN- @hestabbedhim

purdueee Katılım Kasım 2018
2.3K Takip Edilen1.7K Takipçiler
Sabitlenmiş Tweet
ishan
ishan@usrbinishan·
would you really rather choose safe things like. restaurant. or. chocolate. or. idk. whatever
English
0
0
6
2.5K
ishan
ishan@usrbinishan·
@adic_inf 😔😔😔😔 how do i help u
English
0
0
1
12
ishan
ishan@usrbinishan·
@vishalmisra @ShriramKMurthi @Hesamation i did read it but it did feel like i wished a human had paid attention to each sentence and whether it carried essential meaning or just seemed to. the start was pretty strong.
English
1
0
3
137
ℏεsam
ℏεsam@Hesamation·
I often think about professors coming up with these claims.
ℏεsam tweet media
OpenAI@OpenAI

Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946. For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids. An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better. This marks the first time AI has autonomously solved a prominent open problem central to a field of mathematics.

English
18
14
368
44.4K
John
John@haskallcurry·
How does interpolation search work? Well it's really simple. It's just like binary search except instead of picking the middle point as the pivot, you use the hash value to linearly interpolate to a much better guess.
John tweet media
English
2
1
38
1.7K
John
John@haskallcurry·
Imagine you have a sorted array of hash digests and you need to find the position of a particular one (if it exists). Just do binary search, right? Nope! Because hashes are uniformly distributed, you can use *interpolation search*, which is O(log(log(n)).
John tweet media
English
10
6
229
14.8K
ishan
ishan@usrbinishan·
@DanielleFong @lightcellenergy pretty good response and it understood exactly what i was saying. when i was doing quantum simulation i was using pytorch which made it trivial to do fft on gpu. depending on what u want to do with this sim its probably fine to stay with current way u have
English
0
0
1
35
Danielle Fong 🔆
Danielle Fong 🔆@DanielleFong·
Claude and I got a live sodium atom simulation! @lightcellenergy I've long wanted to be able to visualize what is really happening when a neutral sodium atom is excited in a flame or a bulb. Classical theory teaches that the single outer electron orbits the nucleus and full shells, and forms a dipole when this gets "wanged" but, physicists have long wondered, what does it look like when the electron gets wanged? is it like ringing a bell? i thought that this was the perfect time to really push what claude.ai is capable of, in terms of building artifacts. grabbed zig to compile wasm and webgpu to make it super performant, and work with it for a bit and ran the REAL TIME-DEPENDENT Schrödinger equation on a 3D grid in a potential in the browser. and you can kick it! what!! it's soo beautiful... this is going on the lightcellenergy.com website overhaul for sure here is the artifact. i don't endorse that it's exactly right, but i feel it is probably spiritually a good sodium simulation. it certainly rings my dipole! claude.ai/public/artifac…
English
10
7
98
8.5K
ishan
ishan@usrbinishan·
@DanielleFong @lightcellenergy nice, does it do the kinetic part (laplacian) by diagonal using using fft? ime euler style integration is not as good
English
2
0
3
115
佐々木恭子
佐々木恭子@kyokotencoro·
水の中に沈めた鏡で、虹をつくる🌈
日本語
201
2.7K
54.6K
2.9M
Matangi, the Mystic
Matangi, the Mystic@BlumenMit·
A sterling engine at the edge of a moon crater would go really hard with a peak difference in temperature of almost 400°C.
English
18
16
1K
41.1K
Luigi
Luigi@RealLuigiDavid·
For the past two months I’ve been building a liquid rocket engine AND test stand as cheap as possible WITHOUT compromising safety, reliability, or quality. In total I’ve spent ~ $1,000 to build a remotely operated modular stand, and robust rocket engine. Fired 6x yesterday!
Luigi tweet mediaLuigi tweet media
English
15
15
254
12.5K
ishan
ishan@usrbinishan·
@HighFreqAsuka can you tell me more? i feel like it wouldn’t to me, so im trying to understand how you feel
English
0
0
0
80
Asuka
Asuka@HighFreqAsuka·
My team at HRT is hiring! We're building the best LLMs in the world for trading, from scratch. One of the few places in the world you can do real frontier LLM research, from pretraining to reasoning, powered by our incredible B200 cluster. Apply below!
Iain Dunning@iaindunning

New role! We're adding people to our LLM team, ASAP. We already build some of the best models of markets. For LLMs, one must ask in 2026, why another one? We think we have a (unique) story, and are hiring for the full training stack - pre-to-post. Big resources, big ownership.

English
48
29
1.5K
210.1K
ishan
ishan@usrbinishan·
@Kaju_Nut cool argument. also if pedro had the same clock you’d see ur own clock get a billion ticks for each tick of his clock. this is a good framing too since we directly contrast 1 and 1 billion
English
0
0
1
36
Nirmalya Kajuri
Nirmalya Kajuri@Kaju_Nut·
He sees me make my claim and thinks I am crazy. The duration between 2 clicks between my clock is, according to his watch, not that short an interval. This is what Einstein taught us: time is relative. Pedro sees my clock run a billion times slower than I do.
English
4
1
339
28.3K
ishan
ishan@usrbinishan·
@quantum_geoff @SierraHotel28 @crimebucket one possible extension is to say that recursive is bad because stack overflow so now find a non-recursive way so you change to using a stack data structure and maybe later a queue instead
English
0
0
0
24
Geoff Penington
Geoff Penington@quantum_geoff·
@SierraHotel28 @crimebucket I’m not trying to be obtuse here but I only see two possible scenarios for how this goes: 1. They write the required ~2-5 lines of code and you move on to the next question after 30 seconds 2. They say “what does that mean?” or “how do binary trees work in [language] again?”
English
4
0
0
121
Hunter C Davis
Hunter C Davis@huntercoledavis·
I'm excited to introduce Mark 1, Until's electromagnet for rapidly rewarming organs from cryogenic storage temperatures. We set out to scale magnetic nanoparticle warming to human-scale organs while keeping the coil losses low enough for hospital use. Mark 1 produces a 900 kHz oscillating field at up to 40 kA/m with 7 kW of coil loss in a bore compatible with human-sized organs. This intersection of field, frequency, coil volume, and efficiency makes the system a first of its kind. For organ rewarming, we reduce the field strength to 29 kA/m and bring the losses down below 4 kW. At these settings, we have been able to rewarm a vitrified pig kidney loaded with our in-house nanoparticles at 50 C°/min while consuming a total of less than 4 kW out of the wall. At this power, the device could run off of a common single phase 208V circuit. This 50 C°/min warming rate exceeds the critical warming rate required to outrun ice formation using our in-house cryoprotectant. Mark 1 is just one example of the significance of audacious engineers and applied physicists for progress in reversible cryopreservation. If you're excited by pushing the frontier in hardware to improve human health, join our team.
Until@untillabs

We’re putting the “reversible” in Reversible Cryopreservation 🧡 Our latest post breaks down how our rewarming process works and shares new human organ-scale rewarming data we’re excited to finally show. Check it out in the link below 👇

English
6
14
84
7.6K
安藤 しずおかAL330/JS2LLQ
安藤 しずおかAL330/JS2LLQ@AL33012·
国際宇宙ステーション #ISS からの電波、三重県津市から液体アンテナでも取れました！ CMの都合で高度の低い18:30頃までの受信でしたが51でしっかり復調しました👍 コップの水位は14センチ位で2m のλ/4には満たないですが波長短縮効果である程度のマッチングにはなっていたと思われます♪
日本語
136
898
7.8K
843K
ishan
ishan@usrbinishan·
@WKCosmo how does that happen? was something wrong in the original work that concluded that large discrepancy?
English
0
0
0
327

Keşfet

@adic_inf @vishalmisra @ShriramKMurthi @Hesamation @haskallcurry @DanielleFong @lightcellenergy @kyokotencoro