deepsilicon

Software and hardware to run neural networks faster and cheaper

Y CombinatorSummer 2024InactiveIndustrialsEdge Computing SemiconductorsHard TechAI

What they do

Problem

Running neural networks is often too slow and expensive with existing computing infrastructure.

Solution

They build software and hardware that accelerate neural network inference and training at lower cost.

Who it's for

AI developers and enterprises deploying neural networks in production.