Biografía

Disculpa, pero esta entrada está disponible sólo en English.

Beyza is a research assistant and doctoral candidate at IMDEA Networks Institute, and is pursuing her Ph.D. in the Department of Telematics Engineering at Universidad Carlos III de Madrid. She earned her B.Sc. (2020) and M.Sc. (2022) in Computer Engineering from Middle East Technical University (METU), where her master’s research focused on the optimal design of wireless data center networks. She has also been a visiting Ph.D. student at the Networked Systems Laboratory at KTH Royal Institute of Technology.

Her research interests include machine learning-based inference, distributed in-band network intelligence, energy consumption modeling in the data plane, and load-balanced Mixture-of-Experts (MoE) models. She develops systems that enable runtime in-band inference by deploying machine learning models on programmable Intel Tofino switches. She has also designed a system for partitioning large tree-based models and distributing them across multiple programmable switches to address device constraints.

Her work on distributed in-band inference was awarded the Best Paper Award at IEEE INFOCOM 2025. Her current work explores energy modeling of Intel Tofino switches and P4 programs. In parallel, she investigates load-balanced Mixture-of-Experts (MoE) models for improving system throughput.