Subra Seshadri

Title: Subra Seshadri: Innovating in Machine Learning-Driven System Architecture Automation.

Introduction

Subra Seshadri is an inventor based in Broadlands, Virginia. His patented work explores how machine learning and generative models can help automate system architecture tasks.

Latest Patents

Seshadri is associated with 1 patent, titled “Machine learning-driven system architecture automation.” The invention describes a system that receives a user’s query, converts it into an embedding, and selects related artifact embeddings from the same feature space. It then retrieves a knowledge graph whose nodes represent the selected artifacts and whose edges capture semantic relationships among them. The system provides the query representation, selected embeddings, and knowledge graph to a generative model, which produces an output for delivery to the user’s device.

Career Highlights

Seshadri works at Fannie Mae. His patent reflects an approach to combining query embeddings, knowledge graphs, and generative models to support system architecture automation.

Collaborations

Kiran Ramineni and Juliana Carroll are identified as Seshadri’s coworkers. The available information does not specify their roles in the patented invention.

Conclusion

Subra Seshadri’s patent brings together machine learning, knowledge graphs, and generative AI in a system designed to automate aspects of architecture-related work.