Own the machines powering Physical AI. GPU infrastructure you can actually own.
Joined December 2025
- Tweets1.2K
- Following155
- Followers968
- Likes987
Infrastructure needs capital before it needs scale
We are happy to announce that we are partnering with @LiftoWorld to strengthen the path between real infrastructure, serious builders, and the capital needed to grow it
Fundraising Partnership 🤝
@Core_Compute × @LiftoWorld
The point isn’t to put “compute” on-chain
It’s to bring real machines, real usage and real infrastructure into a system that can be owned and verified
That starts right here
The GPUs are only useful while they’re running
Behind the racks are the systems built to keep power stable, bridge interruptions and keep critical infrastructure online
This is the part of AI infrastructure nobody tweets about enough
Owning compute is vague
Owning GPU #147 isn’t
Core Compute is building GPU ownership around specific hardware, not an abstract share of a fleet
If the asset is physical, the ownership should be specific too 💚
Physical AI workloads don’t always fit into an hourly cloud model
Training, simulation and evaluation can need dedicated capacity for weeks or months at a time
30 days, 90 days, 365 days.
For serious workloads, consistency matters as much as raw compute.
More compute density means more heat
And at scale, cooling stops being something around the GPUs instead it becomes part of the infrastructure itself
Power gets the machines running
Cooling keeps them there
AI is leaving the screen
Robots, autonomous systems and embodied agents operate in the physical world but the training, simulation and continuous learning behind them still need serious compute
Physical AI needs a physical layer
That’s where @Core_Compute sits 💚
Everyone wants to see the GPUs
We like showing what keeps them running too
AI infrastructure starts with power 💚
Power continuity beyond batteries.
Core Compute’s data center is supported by a silent diesel generator backup system, providing an additional layer of power resilience beyond the battery/UPS setup.
The system is designed to keep critical infrastructure running during grid outages, voltage interruptions, or extended power failures, helping maintain continuous GPU operations and protect the data center from unexpected downtime.
INSIDE THE HALL / 03
THE WORK STARTS BEFORE THE WORKLOAD
A GPU doesn’t become production-ready the moment it reaches the rack
It gets installed→Burned in→Health-checked→Monitored
Only then is it ready to run real workloads
This is the less glamorous side of AI infrastructure and one of the most important
Welcome inside Core Compute
INSIDE THE HALL / 02
PHYSICAL AI STILL NEEDS A DATA CENTER
Robots move in the real world
But the training, simulation and continuous learning behind them happens here.
With real racks and compute.
This is the Infrastructure for Physical AI
Welcome inside Core Compute
INSIDE THE HALL / 01
WHERE THE COMPUTE ACTUALLY LIVES
GPUs get the attention
But behind every workload is a physical system of racks, power, cooling, networking and infinite operations
Welcome inside Core Compute
This is exactly where the AI infrastructure conversation is heading.
Everyone talks about GPUs. Far fewer talk about what it takes to keep dense compute running 24/7 — stable power, backup systems, cooling, networking and redundancy.
That’s the thesis behind @Core_Compute
We operate 500 enterprise GPUs for Physical AI workloads today, and we’re building the ownership layer around that infrastructure.
As compute density rises, the bottleneck won’t just be access to chips.
It’ll be access to infrastructure capable of keeping those chips working.
The next major AI bottleneck may be delivering enough stable power to keep them running (Save this).
This Goldman Sachs chart maps the companies supplying power and energy storage equipment for next generation AI server racks.
It focuses on battery backup units, or BBUs, and capacitor backup units, or CBUs.
AI workloads can cause a server rack’s power consumption to change within milliseconds and as AI racks become more powerful, these sudden changes can destabilize voltage, create GPU errors or restart an entire system.
BBUs place batteries close to the servers so they can provide electricity during outages, voltage drops or sudden increases in power demand.
CBUs use capacitors and supercapacitors to react even faster and stabilize short power fluctuations.
Now, here are some of the companies that could benefit from this
Panasonic, Samsung SDI, LG Energy Solution, Murata and TDK can benefit from rising demand for battery cells and BBU components.
Delta Electronics, Flex, Lite-On, Advanced Energy, AcBel Polytech and Chicony Power supply power systems, converters or complete backup units.
Texas Instruments, Infineon, Renesas, Analog Devices, NXP, STMicroelectronics, ROHM and Monolithic Power Systems provide the chips and controllers that manage electricity inside AI racks.
Vertiv, Schneider Electric, Eaton, ABB and Legrand sell racks, busbars and power distribution equipment.
Nichicon, Nippon Chemi-Con, Yageo and Rubycon can benefit from growing capacitor demand, while Panasonic, Eaton and Musashi Seimitsu have exposure to supercapacitors.
Every new high density AI rack requires more batteries, capacitors, converters, power supplies, controllers, cooling equipment and electrical infrastructure.
Goldman Sachs reportedly expects the BBU market to grow from approximately $2.2 billion in 2025 to $50.2 billion by 2030.
I’m increasingly focused on the companies sitting behind the power systems inside AI racks, because this bottleneck gets bigger as compute density rises.
If you want to see how Milk Road Pro members are positioned around power, cooling and the broader AI infrastructure buildout, check out my portfolio below.
link.milkroad.com/d8cj6v
Before anything goes on-chain, it has to exist off-chain
Real machines on real racks with real workloads
This is the physical layer behind @Core_Compute
AI needs somewhere to run & these are the machines doing the work.
Core Compute lets you own a GPU from the fleet on @solana while the hardware stays in the data centre running Physical AI workloads
The rack stays here and the ownership goes on-chain
A GPU isn’t “live” the moment it reaches the data center
First it gets racked then burned in and then health-checked
Only after that does it become revenue-eligible
The less glamorous side of AI infrastructure is usually the part that matters most 💚
A robot might spend 1 hour in the real world but before that, it can spend hundreds of hours learning in simulation.
That’s the part of Physical AI people underestimate.
Robots don’t just need compute when they move.
They need it long before they ever touch the real world.
Compute is the real power
