Perforated is the data-efficiency layer for machine learning that helps teams build smaller, more accurate models using less training data. By adding biological-inspired dendritic modules to artificial neurons during training, Perforated allows PyTorch models to extract more value from existing datasets.

Perforated launched in 2025, and is headquartered in Pittsburgh, Pennsylvania. We also have an office in San Francisco.

Today's AI is built on an outdated model of biology. For 80 years, AI has relied on a foundational mathematical model that simply sums inputs. Modern neuroscience has revealed that dendrites, and not neurons, are the core computational unit of biological brains. Perforated AI's artificial dendrites emulate the non-linear integration and processing capabilities of biological dendrites, allowing artificial neuron modules to perform complex computations, which fundamentally modernizes how AI learns.

Perforated is led by CEO and neuroscientist Dr. Rorry Brenner, who pioneered the dendritic approaches to next-generation AI infrastructure. The Perforated team consists of machine learning experts and technologists with presence in San Francisco, Pittsburgh, Boston, Austin, and beyond.

Perforated is the learning-efficiency layer for AI. Our mission is to accelerate the transition from brute force AI to parameter-efficient AI. Our dendritic learning technology enables additional learning signals, helping models achieve higher accuracy with fewer parameters, less data, and lower deployment costs.