Zürich, Switzerland

Foteini Strati

This inventor holds 1 USPTO granted patent. Top assignee: Microsoft Technology Licensing, LLC. Active years: 2026.


Average Co-Inventor Count = 1.0


Company Filing History:


Years Active: 2026

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Foteini Strati: Innovator in Generative Model Serving

Introduction

Foteini Strati is a prominent inventor based in Zürich, Switzerland. She has made significant contributions to the field of generative models, particularly in enhancing their performance and fault tolerance. Her innovative approach has led to the development of a unique patent that addresses critical challenges in model serving.

Latest Patents

Foteini Strati holds a patent titled "KV-cache streaming for improved performance and fault tolerance in generative model serving." This patent describes a method for serving a generative transformer model, which includes determining an optimal batch size for processing inference requests. The method allocates prompt and token pipelines to efficiently handle these requests. The design of the pipelines is influenced by various factors, including batch size, average prompt length, cache requirements, and memory footprint. The use of cache streaming allows for effective token generation through gather-copy operations, enhancing the overall performance of the model serving system.

Career Highlights

Foteini is currently employed at Microsoft Technology Licensing, LLC, where she continues to push the boundaries of technology in her field. Her work is characterized by a commitment to innovation and excellence, making her a valuable asset to her team and the broader tech community.

Collaborations

Foteini collaborates with talented individuals such as Amar Phanishayee and Jakub Michał Tarnawski. These partnerships foster a creative environment that drives forward-thinking solutions in technology.

Conclusion

Foteini Strati's contributions to generative model serving exemplify her innovative spirit and dedication to advancing technology. Her patent reflects a significant step forward in improving the efficiency and reliability of generative models.

This text is generated by artificial intelligence and may not be accurate.
Please report any incorrect information to [email protected]
Loading…