Faith | Pits to Pixels | Power | Compute | Bitcoin
A forest
Joined November 2008
- Tweets104K
- Following2.9K
- Followers13.3K
- Likes68.4K
Patrick Rooney retweeted
Could not be more excited to announce Spark Liquid: @CrusoeAI’s newest modular datacenter product manufactured in our newest factory.
- builds on the proven Spark platform, which is the only modular datacenter platform to achieve NVIDIA Exemplar reliability and performance standards
- deployable in weeks not years
- compact and low-profile
- compatible with any energy source
- massively scalable
- zero operational water use
- designed for next gen @nvidia chips and DSX-compatible
New: Crusoe Spark™ Liquid, our liquid-cooled modular AI factory built for @NVIDIA Vera Rubin NVL72 rack-scale systems.
5x the compute density of air-cooled Crusoe Spark, and part of the NVIDIA DSX POD design partner ecosystem.
Plus: Managed RL and Managed MLflow in private preview ⤵ crusoe.ai/resources/blog/cru…
Patrick Rooney retweeted
Delaying compute futures has a real cost for US innovation. Some of the largest data center lenders have hit their risk limits. Give them a liquid market to hedge compute price risk and they can free up lending capacity. It's the lack of a hedging instrument that has caused credit spreads to blow out.
We support the @CFTC’s commitment to getting this right. The good news is we already have standards for building these markets properly: independent benchmarks, transparent methodologies and regulatory access to the underlying data. Apply them to compute and let well-designed futures move forward.
Our comment letter: regulations.gov/comment/CFTC…
Power at the forest edge isn’t a side bet on the compute trade.
It is the trade.
Chips sit idle in warehouses. Grid queues stretch for years. Bitcoin miners already spent a decade learning the hard part: show up where the electrons already are. Hyperscalers are still catching up.
Attended a great seminar hosted by @yesenergy today at Nodal Trader Conference. I found this map to be curious. All the dots represent demand growth for power & data centers. Why is Silicon Valley, the tech capital of the world, a desert here?
Patrick Rooney retweeted
SO much of what passes for human performance advice amounts to: how do I avoid or get past an unpleasant feeling asap? Or any feeling for that matter! How to not have your feelings is the underpinning of most performance advice.
Maybe the performance advantage is in being able to feel and understand it instead. (Ok, skip the maybe part.)
Hoarding physical compute has a myriad of problems written all over it.
Business Insider confirms it: Amazon has warehouses of GPUs that can’t be installed, and is buying them so far in advance that a supply chain team said they “didn’t have a clear forecasted path” to installing them.
businessinsider.com/amazon-2…
Patrick Rooney retweeted
🚀🚀🚀🚀🚀
Elon Musk is our current Da Vinci.
We need more geniuses like him.
SI + dozens of geniuses will make the world a paradise.
Patrick Rooney retweeted
I was listening to an X space yesterday about AI, and the prevailing theme across 4-5 speakers I listened to was doomerism.
What’s more, nothing of what they said matched up with my experience of reality.
I couldn’t help but wonder if these types of conversations were had whenever other kinds of revolutionary technologies had been released in the past, such as the wheel, printing press, etc
It also struck me that the speakers had no idea what the future held despite positioning themselves as AI authorities.
Their opinions were ripe with guesswork and anecdotes. Which then led me to ask, “if you’re guessing what the future holds anyway, why take the pessimistic doomer version and focus your energy and conversation around that?”
Didn’t the wheel and the printing press work out?
⚡️Power at the node is not a side bet on the compute trade. It is the trade. Everything else—chip
generation, contract tenor, index methodology—is secondary to the question of whether the power
hedge matches the physical location of the load. 🔌
Why was the CBOE building chosen as the location for a data center? Are there really that many traders left in the loop hungry for proximity? Likely not. It was the infrastructure.
Thirty-three megawatts downtown with no major grid upgrades required is only possible because CBOE spent decades building out that electrical infrastructure for trading floor density. The building’s most valuable asset was never the real estate — it was the wiring. That’s a story about how legacy infrastructure becomes the bottleneck and the moat at the same time, and it ties directly into the broader thesis that power, not chips, is the scarce input for AI.
There’s ‘a more to the story. Check out the latest Pits To Pixels: patrickrooney.substack.com/p…
Human filters…their gut…cannot be coded.
As AI makes execution cheaper, we believe that a company’s edge will come from human judgment (which makes hiring the right humans more consequential than ever.)
Proud to be backing the tryprism.com team, which is putting a dedicated talent team within reach of every founder.
My list here on X for all things power markets is talking of nothing other than the Tesla Roadster this evening. It's easy to understand why. $TSLA
Meta's powerful new AI model can ditch the data center and run locally from a desktop foxnews.com/live-news/ai-use… #FoxNews @FoxNews $META
