Akash Dhasade
Akash Dhasade

Postdoctoral Researcher
Carnegie Mellon University

Email: adhasade [at] andrew.cmu.edu Office: Collaborative Innovation Center (CIC), CMU

About Me

I am a postdoctoral researcher at Carnegie Mellon University, working under the supervision of Prof. Gauri Joshi. I completed my PhD at EPFL advised by Prof. Anne-Marie Kermarrec. Before that, I obtained my bachelors in Computer Science & Engineering at the Indian Institute of Technology Tirupati.

My research interests lie broadly in efficient machine learning systems for training and inference. During my PhD, I worked on the challenges of communication efficiency [NeurIPS'24], robustness [ICLR'26], privacy [IPDPS'22], and unlearning [MIDDLEWARE'24] in federated and decentralized learning. More recently, I have been working on making large models practical to deploy and train: merging models with tunable accuracy–size trade-offs [ICLR'26] and jointly tackling privacy–utility–scalability in federated training of billion-parameter models [NeurIPS'26].

Recent News