Mukund Madhav Varma

Hi, I’m Mukund. I’ve spent the last decade or so building AI/ML for precision medicine, first as the person writing the models and now mostly as the person running the teams that do. Along the way I’ve worked on everything from single-cell data to drug programs that made it into the clinic. More on LinkedIn.

I am Director of BioML at the Ideker Lab and Head of Machine Learning for the ARPA-H funded Dynamic Digital Tumors (DDT) program. I also advise and consult for Mycsology Foods, a food technology company developing the next generation of sustainable food ingredients.

Before UC San Diego, I held technical leadership roles in the biotechnology industry. I was Principal Machine Learning Scientist at Caris Life Sciences and computational lead for the anti-TREM1 program at Celsius Therapeutics. I trained at the Broad Institute and hold an S.M. in Physics from MIT and a B.Tech from IIT Bombay.

I have worked on computational problems in

• interpretable deep learning in molecular biology, including foundation models
• early discovery in therapeutics: target discovery, precision medicine and biomarker discovery
• omics data analysis for cancer, immunology and inflammatory disease, with deep expertise in single-cell RNA-seq
• biomarker and risk prediction with computer vision models on histopathology slides
• deployment of models, pipelines and web applications in cloud environments

My computational areas of expertise include

• statistical methods used in bioinformatics
• machine learning model development, classical and deep learning
• test-driven software development in cloud environments

The best way to reach me is through LinkedIn.