Something New prev vc @immutable @Delphi_Digital
New York
Joined January 2021
- Tweets1.4K
- Following4.4K
- Followers434
- Likes3.5K
some of these earnings are not even that bad, and getting punished super hard
just tells us about the state of things and soo much growth being priced that minor hicup is causing cascading reactions
this makes sense
also the price next door impact your price, regardless of the utilization
A lot of ink has been spilled on GPU financing lately. I spent over a decade trading ABS on Wall St, and structured over $2B+ of debt financings as an operator (including at @tryramp). Figured I'd try to make sense of this market, so I went back to a framework I knew well: aircraft ABS and aircraft engine ABS.
My take: a GPU neocloud runs on airline economics, and the debt behind it looks a lot like aircraft engine ABS.
The parallels:
• Perishable inventory. An empty seat after the door closes is gone forever. So is an idle hour on an energized H100 rack.
• 85%+ of cost is fixed (servers, 10-year rack leases, 10-14% debt service). One more hour of compute costs pennies of power.
• Utilization is everything. Airlines break even around 80% load factor, a levered GPU cluster around 65-75%. The returns live in those last 25 percentage points.
• NVIDIA is Boeing and Airbus rolled into one. Powered shells and grid interconnects are the new airport slots.
• Wall St copied the aircraft financing playbook: take-or-pay contract plus servers in a bankruptcy-remote SPV, 70-80% LTV.
Part 1 of my blog post goes into a lot more detail:
alexyangsong.com/blog/alumin…
Yash Gandhi retweeted
Remember I was just saying they would expand products the other day? @AtlasShrug1 $MXL
MaxLinear is making a serious push into the active electrical cable (AEC) market, directly challenging Credo Technology (CRDO), the dominant incumbent.
The most interesting development came from CEO Kishore Seendripu during MaxLinear's Q3 2026 earnings call, held just two days ago.
He outlined an ambition to capture 20–30% share during the 1.6 Tbps AEC generation over the next couple of years, initially targeting one or two major US data-center customers. When it comes to constrained capacity, unlocking Samsung production capacity is a real edge.
the questions for onchain options platforms is, why would anyone use them over robinhood or another local platform which has better liquidity and much better ux
for coins it is a no brainer
for stock, i think off hours or access to offshore assets(relative to your location)
Yash Gandhi retweeted
2 hours ago, Anthropic filed for an IPO.
We went through the S-1 and broke it down in 6 charts:
all things that i would use a computer for works perfectly, but the usual phone tasks seems challenging
13 hours my my new iPhone Duo, picked up at the Apple Store in Piazza Liberty In Milan as first people on the queue as every year, and here is my whole day impression:
- if you knew for years you wanted one this is all u wanted as a first gen
introduction: i own e eye ipad size: mini, 11 with nanotexture, 13 - and the ipad 12 was my preferred one to use every day for the lightness and portability but i always dreamt for the nano to get an oled screen with higher refresh rate and i would have stopped using an iphone
well, the duo is the answer to my dream of a good ipad mini but with the day to day use of a iphone.
> the small display, perfect size now my max looks too long and it only better to write with e ough real estate over the duo, but the more rectangular aspect ratio won me over for being good and big enough
> once the duo is open, i forget to close it back until i want to move away, unless im watching a video.
>> videos from youtube are just amazing, whole screen, perfect t aspect ratio.
>> safari pages are finally real desktop app, another quality of live upgrade i love about ipads that is finally on the phone in la escape mode anywhere and in portrait on many
> apps, they are already getting updated as im writing, already 15 apps i use got a duo upgrade within the day, bullish on the format, but the small rectangle sucks for the ones still lagging so praying that every dev moves quickly.
but thats why i wanted an apple foldable and not android: developers care and they understand the advantages in positioning themselves quicker than the competition
> battery, it’s still optimizing as first restore so i know it’s not a benchmark, but all my magsafe batteries now are longer than the Duo so i probably have to buy new ones if i want to daily drive the duo at a conference the whole day cause it eats faster than the max for now
> camera is ok, not the best, but we platoed long time ago and upgrading just for the camera a phone is becoming meaningless so im ok with the big upside of being able to take photos looking at myself. will only use the rear camera with duo display mode on and won’t touch the inner one
> it’s light, and touch id is fine. took me some hours to fight the muscle memory but it will probably save me from having to to weird late night acrobatics to get authenticated while in bed so it has pros and cons. but truly fine
> audio is fine, not ipad level but enough to make calls and watch movies on a big display while moving around.
> novelty effect, a full go upgrade it for me.
as an ipad fanboy that can rarely justify why i update my phone every year, this is the device i always wanted and my lifestyle is the one of someone that know how to benefit the hell out of this
> utility 10/10 for a professional that uses his phone to monitor his main business, for someone like me trading at all time, it was a dream that one day we could trade from our palms but the experience was always subpar vs a full desktop environment where you could appreciate a chart much more.
now the gap is filled, and your trading desk lives in your pocket, one iphone unfolding away.
if you are a trader, a vibecoder, or anyone that wants to move their desktop setup to their pocket without having to carry an ipad with them, this is the device you have been waiting for
sick
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
Yash Gandhi retweeted
> can’t own stonks with capex slowing
> can’t own crypto with quantum and ai risk
> can’t browse the internet cause everything is hacked
> can’t own software cause everything is hacked
> can’t own gold cause alchemy is solved
> can’t own bonds as inflation and crowding out hurting it
> mfw
finally we have an app leading infra change
we have always had the cart before the horse
on the investor call Vlad mentioned vanilla option by EOY with
- cross margining b/w options, perps and spot
- regulatory compatibility, opening up US customers
- potential link to Robinhood apps where possible
gg @trevor_flipper
i'll write a longer post later but TLDR:
* Jeff has made it clear vanilla options are a long ways out for Hyperliquid (if ever) and I'm not concerned about Binary options.
* Lighter has shipped Binary options and plans to ship Vanilla options soon. I'm skeptical that will come out on time or be released on that timeline without some serious compromises made to the risk/margin system for those positions (read: inefficient) given the inherent challenges of building this risk engine
Even in the scenario where Lighter ships options in a timely and appealing way - they still have to compete. Derive has 90%+ share of options premiums paid. Obviously that won't stay that high forever, but it shows they have a very serious lead. We've already seen in centralized exchanges that perps venues don't automatically win options markets
Binance & OKX have the most perps volume by a large margin, you'd expect them to also be leading in options but they aren't. The centralized options leaders are two entirely different exchanges Deribit and ByBit. Again, this is because market structure & risk engines are just a different thing for options. So we've seen the leaders in that space keep their lead despite well funded and talented competition. I don't think it's unfounded to believe that at least for the foreseeable future, Derive is in a position to keep it's lead and go toe-to-toe with the leading perps dexes at least in the domain of options.
FINALLY let's assume Lighter does win options. I don't imagine it will win over night. You'll have time to see it coming in the metrics. In that scenario I'm happy sacrificing some of my Derive PNL (bc i held too long) and getting a worse entry on Lighter (bc i bought too late) because Lighter passing Derive with strength implies the market for this will GROW and you could still make a lot of money buying LIT at that moment
neutral
they generated $16.3b in revenue in the first half(~33b annualised) and will spend $750b?
Latest Anthropic S-1 draft shows big uptick in future compute spend!
Anthropic could spent $750 billion in compute versus prior S-1 draft of $518 billion. That’s a lot more chips.
WSJ: “The company has signaled that it could potentially spend $750 billion in the coming years on obtaining more computing capacity, including deals it is still negotiating.”
“Anthropic’s future infrastructure costs top $600 billion. The firm is negotiating additional computing deals that could see its commitments rise by as much as $150 billion, the materials show.”
Vs. Reuters Sept. 28 prior S-1 draft $518 billion wsj.com/tech/ai/anthropic-re…
