optoML has developed an in-memory compute architecture with optical interconnects that are designed to offer higher energy efficiency and lower latency and costs.optoML, a fabless semiconductor startup has raised $1.8 million in a pre-Series A funding round led by Bluehill.vc & a99.
The new funds will be used to boost hiring and help it start work on next-generation chips after a tape-out with Taiwanese semiconductor maker TSMC is completed. In semiconductor parlance, tape-out refers to the final stage of the design process where the design is complete and is sent for fabrication.
optoML has developed an in-memory compute architecture with optical interconnects that are designed to offer higher energy efficiency and lower latency and costs.
“As AI workloads scale from the edge to hyperscalers, energy efficiency, latency, and bandwidth are becoming first-order constraints. optoML’s 12nm tapeout with TSMC and its partnership with Kaynes Semicon mark a critical transition from research to real silicon and scalable production,” said Manu Iyer, general partner at Bluehill.
optoML was founded by Saravana Maruthamuthu who has over 17 years of experience across the likes of Intel, Qualcomm, Continental and Isar Aerospace.
“Bluehill and a99‘s conviction will put us on a path to clear productization, qualification, and real world impact, turning efficient AI compute into deployable systems,” Saravana said.
The startup said its patented in-memory compute architecture offers up to 50 times higher energy efficiency compared to traditional digital accelerators.
“We at a99 are thrilled to invest and be a part of the optoML juggernaut. As AI workloads move to the edge, power efficiency becomes the real constraint. optoML’s analog-in-memory approach is designed to address exactly that,” said Vignesh Shankar, general partner at a99.
Leave a Reply