This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignees: Massachusetts Institute of Technology, Ntt Research, Inc.. Active years: 2026.
Company Filing History:
1 out of 8,573 
1 out of 33 Years Active: 2026
Title: Manya Ghobadi: Innovator in Optical Inference Technology
Introduction
Manya Ghobadi is a prominent inventor based in Cambridge, MA (US). She has made significant contributions to the field of machine learning and optical technology. Her innovative work focuses on reducing latency in machine learning inference through advanced optical processing techniques.
Latest Patents
Manya Ghobadi holds a patent for her invention titled "In-network Optical Inference." This technology provides low-latency machine learning inference by utilizing programmable switches and optical matrix multiplication. The In-network Optical Inference (IOI) system employs a transceiver module known as the Neuro Transceiver, which features an optical processor capable of performing linear operations, such as matrix multiplication, in the optical domain. By integrating these transceiver modules into programmable packet switches, the system can execute non-linear activations in the electronic domain and respond to inference queries. This innovative approach significantly reduces the end-to-end inference latency experienced by users, as it processes queries within the network rather than relying on cloud or edge inference servers.
Career Highlights
Manya has worked with esteemed organizations, including the Massachusetts Institute of Technology and NTT Research, Inc. Her experience in these institutions has allowed her to develop cutting-edge technologies that push the boundaries of machine learning and optical processing.
Collaborations
Manya has collaborated with talented individuals such as Zhizhen Zhong and Weiyang Wang. These partnerships have contributed to her success and the advancement of her innovative projects.
Conclusion
Manya Ghobadi is a trailblazer in the field of optical inference technology, with her patent for In-network Optical Inference showcasing her commitment to innovation. Her work continues to influence the landscape of machine learning and optical processing, paving the way for future advancements.