Teams at d-Matrix

Develop and optimize runtime firmware and low-level drivers for a multiprocessor SoC used in AI inference. Architect, implement, and validate system software, collaborate with hardware and verification teams, ensure runtime performance and delivery schedules, and debug distributed/scale-out systems.

Lead SoC architecture definition and delivery for AI accelerators. Analyze AI workloads, drive hardware/software co-design, performance modeling, simulation tools, kernel and networking fabric work, port and optimize customer workloads, and collaborate across RTL, DV, and software teams.

Design and implement efficient algorithms to optimize LLM inference on DNN accelerators. Collaborate with mathematicians, ML researchers, and engineers to apply advanced algorithmic and numerical techniques for high-impact generative AI performance improvements.