D-Matrix AI Chip Promises Efficient Transformer Processing

Startup combines Digital In-Memory Compute and chiplet implementations for data-center-grade inferencing.

This article was written by Cambrian-AI Analysts Alberto Romero and Karl Freund
D-Matrix was founded in 2019 by two veterans in the field of AI hardware, Sid Sheth and Sudeep Bhoja, who previously worked together at Inphi (Marvell) and Broadcom. The company was born at a singular moment for the field of AI, just two years after the popular transformer architecture was invented by Google Brain scientists. By 2019, the world was starting to realize the massive significance of transformer-based models and D-Matrix saw an opportunity to define its AI hardware specifically to excel in using these Large Language Models.

Read the full article on Cambrian AI Research

Suggested Articles

documentation AI

IA for AI: The New Reader That Doesn’t Use the Nav Bar 

By Roopa Malavally, Director of Technical Documentation | July 14, 2026

Where heterogeneous pipelines power an agentic coding future

By Matthew Lynley | May 14, 2026

The new perf/TCO math: maximizing GPU utilization with disaggregated pipelines

By Matthew Lynley | July 28, 2026