Building Learning Machines
Know your goal
Read more →
Rohith Krishna, PhD
Computational Biology & Machine Learning
I build models of biology—from atomic coordinates that form molecular interactions, up to the organization of pathways in cells. These models have been used across the world in both industry and academia.
Previously: Baker Lab (UW) • Denali Therapeutics • UC Berkeley
Led development of three foundational models that enable AI-powered molecular design:
Developed ML infrastructure and analysis tools for therapeutic discovery at Denali Therapeutics over 2 years.
Worked directly on drug development programs, processing millions of sequences and generating predictions for cross-functional teams.
Understand the practical challenges of commercial biotech.
Build complete ML systems for molecular design: from model architecture and training infrastructure to experimental validation and real-world deployment.
Expertise spans developing custom PyTorch model code to experimental biochemistry.
Know your goal
Read more →This was written as a reflection for BIOC 541. I have no intentions of conducting this study, if you do and have cool results, let me know!
Read more →TLDR: I trained a neural network to predict how much area of each amino acid in an antibody structure can interact with other molecules. The predictor worked really well for...
Read more →