I am currently working as a Machine Learning Engineer, focusing on building fast image processing pipelines using deep learning. My work involves training, testing, and deploying vision models for high-resolution image matting, color correction, and enhancement.
Machine learning excites me as a field because doing extensive experimentation to find out what works and what doesn't, and then developing reasoning about why something might be working, makes me feel like a spy trying to solve a mystery or a physicist trying to understand the nature of nature.
I am broadly inquisitive about large language models, vision transformers, PEFT (parameter efficient fine-tuning), and low resource learning. I love learning CUDA and playing around with kernels.
I am always open to discussing ideas and learning new perspectives. If you are interested in collaborating or have any questions, feel free to reach out on any platforms mentioned below!