Vijay

jacobian conjuncture is pretty useless it seems.. just pure maths stuff. so no science or material or energy breakthrough.

@ProteanEgovTech Received e-pan. Thanks. You can close the case.

@ProteanEgovTech Not addressed. I was asked for more info. I've provided more info. Nothing solved yet.

.@ProteanEgovTech Good morning. I'm following up for my e-pan for the last 10 days. check the tx reference send me the pdf.. it should take less than 5 mins. Tx reference 20260707133807000856 Time 2026-07-07 13:38:41.0 Amount paid ₹8.26

@__alpoge__ gemini, chatgpt, ds agrees to the proof but llama pushes back

@__alpoge__ jacobian conjuncture was proven false? gemini, chatgpt, llama, ds: no, not yet claude: yes, it was proven

hello there the jacobian conjecture is false thanx to my close friend akhil for asking about it and my other close friend fable for working during the world cup final ((1+xy)^3 z + y^2 (1+xy) (4+3xy), y + 3 x (1+xy)^2 z + 3 x y^2 (4+3xy), 2 x - 3 x^2 y - x^3 z): \C^3\to \C^3, has jacobian determinant -2, and sends (0, 0, -1/4), (1, -3/2, 13/2), and (-1, 3/2, 13/2) to (-1/4, 0, 0)

@ProteanEgovTech I received a call from ezee will. I've shared info asked to your dm, as requested. Call me on the same number, shared in you DM, in case you need more info.

@ProteanEgovTech 2 weeks, no response yet

@Polymarket wow, this guy alone increased the ai bubble burst likelihood few percentage points?

16% chance the AI bubble bursts. poly.market/tkLeR5o

NEW: San Francisco woman claims her husband handed her nearly all parenting duties so he could become “AI native” — spending days, nights, & weekends locked in his office learning AI.

@che_shr_cat Intelligence is fractal.. MoE, then within experts there are effective weights, then test time scaling, looping/ReAct, GPUs doing 90% tokens at 50% power, etc.

1/ Parameter scale is a brute-force crutch. What if a 35B model could beat a 1,000B model simply by scaling its search horizon at test time? It is not magic. It is structured process feedback. Let us look under the hood. 🧵

@stats_feed standing for just 3-4 hours a day for five days a week is the physical equivalent of running.. 10/365, 2.74% marathons a day, or 1.15 km run a day

Standing for just 3-4 hours a day for five days a week is the physical equivalent of running ten marathons a year

@ns123abc this person looks like fit, well-fed version of vitalik

BREAKING: Google is developing a specialized inference chip because the compute crunch is so bad Google Cloud is turning down deals New chip line: "Frozen v2" — descended from a Jeff Dean design that proposed etching Gemini weights directly into silicon v2 freezes the architecture instead: >gemini architecture hardwired into the chip >10x tokens per WATT than newest TPUs for inference >deployable as soon as 2028 DeepMind is betting that the transformer architecture is done evolving The model is the chip now

@SemiAnalysis_ they make 18a krin but ascend only 7nm?

We cut open the Kirin 9030 and put it under an electron microscope. The smallest metal pitch measures 32.5 nanometers. That is tighter than Intel 18A, their brand new leading edge node. A Chinese fab with no EUV is out-pitching Intel's EUV node by roughly 10 percent. What's going on here? 🤯 "This is the HiSilicon Kirin 9030, the chip inside Huawei's newest flagship phone. A few weeks ago, we cut it open, put it under an electron microscope, and measured the smallest wires inside the chip, the metal pitch. And what we saw was unexpected. The smallest metal pitch inside the Kirin 9030 measures only 32.5 nanometers. That’s smaller than the metal pitch in Panther Lake, which is based on Intel's brand-new 18A node. A Chinese fab, cut off from the most advanced tools, without EUV, is packing its wires about ten percent tighter than Intel's leading edge EUV node. What’s going on here?"

@KobeissiLetter frozen v2

BREAKING: Alphabet, $GOOGL, is planning on launching a new “frozen” chip to run its AI models more efficiently, per The Information. Details include: 1. This new server chip would directly integrate the blueprint of its Gemini AI model 2. The chip would then enable Alphabet to provide its AI models to users much more efficiently 3. The chip's developers estimate that it would be 6 to 10 times more efficient than the most recent version of Google's existing line of AI chips Another tech giant has entered the chip race.

what kind of risk capital do they have in ap/ts?

10cr liquor scam.. period film :)

Watching #MaaIntiBangaaram Half the way through.. not met expections.

half ass attempt. neither informative, nor insightful, nor funny. "i've never bought bitcoin". that's genius, @nikhilkamathcio. 2 extras added at the end are ultra useless. @brian_armstrong @ycombinator obv eth gang in btc clothing. youtu.be/p8O6SFwXqok

@jpthor customer: how much? tokenner: that will be 20m tokens

I'm sorry, but I don't think people will use money into the future. it's gonna be tokens. and there will be an infinite supply of them. For the first time in 15 years I've become a bitcoin bear.

@ActusDei banks should take less.. investors profit can't go negative before banks cut goes 0.

India's FCNR scheme is falling short, and that gives NRIs some bargaining power The FCNR(B) scheme (Dollar Deposits from NRIs), RBI's go-to for rupee support since the 2013 taper tantrum has mobilised just $3.08 billion so far from disclosed numbers. Original estimates pegged this scheme at $50-70 billion. That's not a shortfall, that's a scheme running at a fraction of its target with the window closing September 30, 2026. For NRIs, that gap is leverage literally and figuratively. Some points to note: → Only a handful of banks have disclosed numbers, and each has a long way to go to achieve their target. PNB: $425mn raised against a $2.5bn target. Indian Bank: $140mn against $2bn. Union Bank: $106mn against $2bn. If a bank is telling you its target, it's also telling you how much room it has left to fill. → HDFC, ICICI, Kotak, Federal, and Yes Bank haven't disclosed their numbers publicly. That doesn't mean the appetite isn't there. It means you have to ask and compare, bank to bank, rather than assume one is offering the best deal. → Leverage on these deposits ranges from 8-12x at Indian banks, up to 19x at some overseas institutions, a meaningful difference depending on where you bank. → The leverage math isn't universal. FCNR interest is taxed on a gross basis for most NRIs, and per HDFC Bank's own CEO, leveraged positions mainly pencil out for depositors in the Gulf and Singapore. If you're in the US, UK, or Australia, the tax treatment in your country of residence changes the calculation, even though the interest itself is tax-free in India. The bottom line: banks are competing for NRI dollars right now, with roughly ten weeks left on the incentive window and the system nowhere near its mobilisation target. Most depositors don't realise how much negotiating room that gives them. We will guide you through this, including the risks to watch out for. Schedule a call with us on thefynprint website. Sources: Hindu Business Line, Mint, ET HDFC Bank management commentary. #NRI #FCNR

@JasonPLowery 20 GW @ 1000 EH/s, 20 j/TH

Bitcoin is worth ~$64k, but it costs ~1 GW. Understanding that difference is everything.

@Mandrik I remember baklava, socks stories :)

My excitement around Bitcoin is gone, but the love remains after nearly 15 years. Maybe a future innovation will spark that fire again. Maybe I'll do the baklava man gimmick forever, like a band that doesn't know when to quit. Maybe I'll randomly vanish one day. One thing I'm sure of - Bitcoin has always done what I needed it to do. Maybe that's good enough, no matter what the future holds.

@lithos_graphein imagine 2x 600 mm water scale chip folded on top of one another

Wtf?