Univ. of Washington machine learning spinout OctoML raises $85M to expand its partnerships

OctoML CEO Luis Ceze. (Photo courtesy of Luis Ceze)

New funding: Machine learning startup OctoML has raised $85 million in new funding, the Seattle-based company announced Monday. The fresh cash brings the company’s total amount raised to $132 million, including a $28 million round in March.

Maximizing machine learning: OctoML uses machine learning to help organizations build machine learning and AI models and applications. The University of Washington spinout helps optimize and fine tune performance of such applications.

“The ‘secret’ sauce in our technology is to use ML to optimize ML, reducing the optimization and tuning time by orders of magnitude,” said CEO Luis Ceze in a previous blog post. Ceze is a UW computer science professor who co-founded the company in 2019. The OctoML platform is built on an open source framework developed by Ceze and his colleagues, called Apache TVM.

Powering up partnerships: The company aims to grow its partnerships with hardware and cloud vendors to extend the use of its tools to different types of uses. In the last month, the company has announced partnerships with Qualcomm, AMD, Arm and other hardware vendors. OctoML’s software is also used by Microsoft, Bosch, Toyota and many others.

RELATED: Just call it deal mania: A look at this year’s booming market for M&A, IPOs, SPACs and VC

“Our ecosystem efforts are driven by our vision for the company, which is to make ML accessible to as many developers, anywhere and on any device,” said Ceze in a statement. The company says its users achieve a two- to ten-fold increase in performance of their machine learning models.

The financing will enable OctoML to expand its lead machine learning deployment platform and build and expand its partnerships.

The backers: The Series C funding round was led by Tiger Global Management with participation from existing investors Addition, Madrona Venture Group and Amplify Partners.

Startups Funding Luis ceze Machine learning Octoml University of washington