Hello, welcome to my home page!
I currently work at SambaNova Systems as a software engineer looking at improving large language models (LLMs). Currently, I am researching alignment methods to make language models comply with enterprise user preferences, which include techniques like RLHF, preference model pretraining (PMP), RLAIF, etc. Another aspect that I am working on is the viability of using State Space Models (SSMs) as an alternative to attention-based transformer models for language tasks on the SambaNova Systems software stack. I am especially interested making models more effiencent via sparsity either by reducing computational burden during training or inference with structured sparsity.
I did my PhD from UW-Madison with Mikko Lipasti. You can click the About Me tab to learn more about my background or visit my linkedin profile.
My goal for this upcoming year in 2024 is to create a set of blog posts summarizing papers and explore interesting ideas in NLP, as well as start to be more involved in the open-source community. Currently, my website/blogs are still under construction but there should be more interesting content coming up soon!
If you would like to get in touch, please contact me via email or through linkedin!