Max Robotics — Global Service Robot Directory + US Certification

About

Viansa Schmulbach is a rising star in computer architecture, whose work bridges the gap between efficient machine learning hardware and real-time robotics. Her research focuses on designing novel system-on-chip (SoC) architectures that accelerate transformer-based language models and robotic control loops, with a particular emphasis on near-memory and near-core computing. Schmulbach’s most notable contribution is the introduction of NeCTAr (Near-Cache Transformer Accelerator), a 16nm heterogeneous multicore RISC-V SoC that achieves 109 GOPS while running at just 400MHz and 0.85V. This prototype chip, detailed in her 2024 paper, demonstrates how sparse and dense matrix-vector multiplications can be performed efficiently by integrating both near-core and near-memory accelerators. Her work has already garnered 5 citations in its first year, signaling strong early impact in the architecture community. By tackling the twin challenges of language model inference and robotic control on a single, low-power platform, Schmulbach is paving the way for more capable edge-AI systems. Her achievements represent a significant step toward making advanced AI accessible in resource-constrained environments.