Faster execution for AI workloads
Open for applications
Faster execution for AI workloads
101-200 employees
Open for applications
101-200 employees
To help companies run AI workloads to run neural networks faster, cooler, and cost-effectively using at-memory computing.
Jump to section
To help companies run AI workloads to run neural networks faster, cooler, and cost-effectively using at-memory computing.
Increased use of AI, and the hardware requirements that come with it, have presented a challenge for traditional datacenter compute architectures. Untether AI was launched to help companies run their AI workloads much faster, smoother and without breaking their budget.
The company has developed a unique at-compute chip architecture for neural net inference that eliminates the data movement bottleneck that costs energy and performance in traditional architectures. By having the compute element directly adjacent to memory cells, it accelerates inference for diverse neural networks such as vision, natural language processing and recommendation engines.
With considerable funding at its back, Untether AI is committed to driving customer expansion and continues to develop its next-generation of chip and products, as well as expand its team and partnerships.
Freddie
Company Specialist at Welcome to the Jungle
Jul 2021
$125m
SERIES B
Jul 2020
$6.6m
SERIES A
This company has top investors
Darrick Wiebe
(Co-Founder)With experience as a Technology Artist, they went on to Co-Found XN Logic and LightMesh.
Raymond Chik
(Co-Founder)Co-Founder and Board Chair at Hepzibah and Kapik Integration.
Chris Walker
(CEO & President)Also Senior Advisor at Cerberus Capital Management. Prior to this, they spent almost three decades at Intel Corporation, including as a VP and GM.