elastix.ai

About elastix.ai

ElastixAI is a Seattle-based technology startup developing a software-oriented platform to optimize the performance and efficiency of large language models (LLMs). The company focuses on the AI inference stage, which is how trained models generate responses. Its primary goal is to reduce the cost and complexity of running these models, whether on cloud servers or edge devices, by offering an adaptable and scalable infrastructure.

Founded in early 2025, the company emerged from stealth with a $16 million funding round led by FUSE, with participation from investors including Catapult, Tyche, Liquid 2 Ventures, and DNX Ventures. The founding team consists of former engineers from Apple, Xnor, Waymo, and Meta. The CEO, Mohammad Rastegari, was previously the co-founder and CTO of Xnor, an edge AI startup acquired by Apple for approximately $200 million in 2020. He is joined by co-founders Saman Naderiparizi (CTO), who led hardware engineering at Xnor and was a senior engineering manager at Apple, and Mahyar Najibi, who also worked at Apple and Waymo. Both Rastegari and Naderiparizi are also affiliate assistant professors at the University of Washington.

ElastixAI's platform is designed to be hardware-agnostic and provide flexibility, allowing customers to configure their systems for specific use cases. This approach addresses a gap in the market for delivering scalable and low-cost AI inference. By creating a software solution that dynamically adapts to different hardware and workloads, the company aims to serve a broad range of clients, from large-scale cloud providers (hyperscalers) to enterprises integrating AI into their operations. The business model may involve licensing its software to chipmakers, cloud providers, and device manufacturers.