Welcome.
I am a Senior Research Engineer with 7+ years of experience bridging the gap between algorithmic research and hardware constraints.
Formerly at Qualcomm AI Research, I specialize in Generative AI efficiency, optimizing LLMs, Mixture-of-Experts (MoE), and neural compression for real-world deployment. My work focuses on turning risky research proposals into practical, high-impact algorithms and building the robust infrastructure needed to scale them.
You can find some of my work here, or check me out on social media:
Find some of my recent publications below.
Published in Transactions on Machine Learning Research.
Co-developed and co-authored a quantization library for efficient large language models (LLMs).
Accepted at TMLR.
Accepted at ICLR 2021.
Accepted at CVPR CLIC workshop 2020.
Accepted at ICCV 2019.
Master thesis. Investigating how conditional variational autoencoder can be used for voice conversion with the goal of improving robustness of automated speech recognition (ASR) systems.
During an internship at the Keiser lab at the UCSF (University of California, San Francisco) I developed this framework to build various architectures of Graph Convolutional Networks, and run them on GPU.
Bachelor thesis, investigating the effect of electrical cortical stimulation on visual attention in humans.
Literature analysis about the mechanisms behind consistency in visual perception across eye-movements.
Thanks for sending a message. I'll get in touch with you soon!
An error occured.