Mueon Corp., a data center hardware startup, has raised $65 million in a Series A round, SiliconANGLE reported on October 15, 2026. The round was led by Maverick Silicon, a semiconductor-focused fund. Participants included Intel's venture capital arm, Oregon Venture Fund, Wellington Capital and IQT, along with other investors.
Mueon was founded last year by former Intel engineers Wilfred Gomes, now chief executive, and Jack Hwang, the chief technology officer. Gomes worked on die stacking at Intel, while Hwang helped develop the techniques used to etch transistors into silicon wafers.
The company's product, Cubelet, departs from the conventional server layout in which core components sit side by side on a motherboard. Cubelet stacks compute and memory components on top of one another. Mueon claims the approach could be up to 10 times more power- and cost-efficient than current hardware. That figure is the company's own and has not been independently verified in the source material.
The efficiency logic is that stacking places components closer together, which shortens the distance data travels. Shorter paths reduce power consumption and can also speed up computation.
Heat is a central difficulty for 3D designs. Processors and memory chips often have different maximum operating temperatures, so a processor running near its limit can overheat memory stacked above it. Heat from the top layer also has to pass through several other components before reaching the motherboard, where many servers shed part of their thermal load.
According to a blog post from investor Oregon Venture Fund last year, Mueon has found a way to address this. The post described a Cubelet as a module that bundles compute and storage with the interconnect, power delivery and thermal management hardware needed to run reliably. It also said the modules can host multiple chip types, including graphics cards and Arm CPUs. No technical detail of the thermal solution appears in the source.
Mueon says Cubelets are tuned for AI and hyperscale workloads. Proceeds will go toward hiring engineers and refining the system design, while the company also builds software meant to let customers run popular AI frameworks on its hardware.
Why it matters: Power and density limits are pressing constraints for AI infrastructure, and a tightly integrated 3D module is one proposed response. Mueon is early-stage, its efficiency claim is unproven in public, and the thermal approach is undisclosed, so the funding signals investor interest rather than demonstrated performance.
TwinStack automated analysis, based on reporting by SiliconANGLE. Check the original for full details.