MemWize is a stealth-mode technology startup focused on improving the efficiency of artificial intelligence inference at data center scale. Backed by venture and strategic investors, the company is pioneering system-level innovations enabled by a new category of system-on-chip (SoC). Its work brings together hardware and software development to address the performance demands of next-generation AI systems.

MemWize’s areas of focus include high-performance GPU kernels, inference frameworks such as vLLM, SGLang, and PyTorch, and compiler capabilities spanning PyTorch, Triton, CUDA, and machine intermediate representations. The company is also developing performance and functional simulators to inform architecture decisions and support hardware-software co-design. Its engineering efforts encompass SoC architecture, compute and data-movement design, design verification, and RTL implementation.

Based in Sunnyvale, California, MemWize is building its team across software, machine learning, compiler, simulation, architecture, and hardware engineering. The company presents its work as a platform-level effort to optimize inference throughput and latency, with roles based onsite at its Sunnyvale location.