Software Engineer @ Meta
Performance Optimization for GenAI | Former ML Researcher @ Huawei
MSc, McGill University
About Me
I am a Software Engineer at Meta, where I focus on optimizing the performance of GenAI models for the MTIA (Meta Training and Inference Accelerator). Previously, I was a Machine Learning Researcher at Huawei Technologies Canada, where I developed data-driven AI-assisted systems for complex networking challenges.
I hold a Master of Science (MSc) in Electrical and Computer Engineering from McGill University. My expertise lies in the intersection of Machine Learning, Speech Signal Processing, and High-Performance Systems.
Experience
Software Engineer @ Meta
Optimizing the performance of GenAI models for the Meta Training and Inference Accelerator (MTIA).
Machine Learning Researcher @ Huawei Canada
Developed state-of-the-art Transformer-based Voice Activity Detection and AI-assisted networking solutions.
Selected Publications
Config-Snob: Tuning for the Best Configurations of Networking Protocol Stack
A protocol tuning solution utilizing historical data and causal inference to dynamically select optimal networking configurations.
Tr-VAD: An Efficient Transformer-Based Model for Voice Activity Detection
A novel transformer-based approach for VAD that balances efficiency and performance in complex environments.
Complex IRM-Aware Training for Voice Activity Detection Using Attention Model
Leveraging attention mechanisms and complex IRM-aware training to improve robustness in speech processing.
Featured Projects
STAM-pytorch
Implementation of Spatio-Temporal Attention Mechanisms in PyTorch for advanced signal processing.