Bharath Ramsundar, Ph.D.

Email: bharath@deepforestsci.com

Linkedin
Substack

Education

  • B.S. Electrical Engineering and Computer Science and Mathematics, 2011
    Departmental Citation (Valedictorian) for Mathematics, Class of 2012
    Highest Honors in EECS and Mathematics
    University of California Berkeley
  • Ph.D. Computer Science, Hertz Fellow, 2018
    Thesis Supervisor: Dr. Vijay Pande
    Thesis Title: “Molecular Machine Learning with DeepChem”
    Stanford University

About Bharath

Bharath received a BA and BS from UC Berkeley in EECS and Mathematics and was valedictorian of his graduating class in mathematics. He did his PhD in computer science at Stanford University where he studied the application of deep-learning to problems in drug-discovery. At Stanford, Bharath created the deepchem.io open-source project to grow the deep drug discovery open source community, co-created the moleculenet.ai benchmark suite to facilitate development of molecular algorithms, and more. Bharath's graduate education was supported by a Hertz Fellowship, the most selective graduate fellowship in the sciences. After his PhD, Bharath co-founded Computable a startup that built better tools for collaborative dataset management. Bharath is currently the founder and CEO of Deep Forest Sciences, which is building an AI-powered suite for drug and materials design and discovery.

Bharath is the lead author of "TensorFlow for Deep Learning: From Linear Regression to Reinforcement Learning", a developer's introduction to modern machine learning, with O'Reilly Media, and "Deep Learning for the Life Sciences". Additionally, he authored "The DeepChem Book" in collaboration with the DeepChem team, published in 2024, and is currently working on "Differentiable Physics: Machine Learning for Physical Systems".

Awards and Honors

  • Hertz Fellowship, 2012-2018
  • Departmental Citation (Valedictorian) for UC Berkeley Mathematics, 2012
  • Highest Honors in EECS and Mathematics, 2012
  • UCRegents’ and Chancellor’s Scholar, 2007

Books

Papers and Patents

2024

Preprint

Open Source Infrastructure for Automatic Cell Segmentation.

Preprint

Self-supervised Pretraining for Partial Differential Equations

Peer-Reviewed Conference Publications

Open-Source Molecular Processing Pipeline for Generating Molecules

Peer-Reviewed Conference Publications

Machine Learning-Driven Predictions for Janus Kinase 3 Protein Drug Effectiveness

Peer-Reviewed Conference Publications

Open Source Fermionic Neural Networks with Ionic Charge Initialization

Patent

Techniques for a cloud scientific machine learning programming environment

2023

Journal Publication

Differentiable Modeling and Optimization of Battery Electrolyte Mixtures Using Geometric Deep Learning.

Journal Publication

Scientific discovery in the age of artificial intelligence.

Peer-Reviewed Conference Publications

Building AI Models of Patient-specific Drug Side Effect Predictions

Peer-Reviewed Conference Publications

Open Source Infrastructure for Differentiable Density Functional Theory

Peer-Reviewed Conference Publications

Score Based Models for Molecule Generation

Patent

Determining suitability of machine learning models for datasets

Patent

Foundation model based fluid simulations

2022

Journal Publication

AutoMat: Automated materials discovery for electrochemical systems.

Preprint

ChemBERTa-2: Towards Chemical Foundation Models

Peer-Reviewed Conference Publications

FastFlows: Flow-based Models for Molecular Graph Generation

Peer-Reviewed Conference Publications

ChemBERTa-2: Towards Chemical Foundation Models

Patent

Differentiable machines for physical systems

2021

Preprint

Differentiable Physics: A Position Piece

Peer-Reviewed Conference Publications

Bringing Atomistic Deep Learning to Prime Time

2020

Journal Publication

AMPL: A Data-Driven Modeling Pipeline for Drug Discovery.

Preprint

SARS-CoV-2 and COVID-19: An Evolving Review of Diagnostics and Therapeutics.

Peer-Reviewed Conference Publications

ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction

Peer-Reviewed Conference Publications

Flow Based Models for Active Molecular Graph Generation

Patent

Interpretability-based machine learning adjustment during production

Patent

Determining validity of machine learning algorithms for datasets

2019

Journal Publication

A guide to deep learning in healthcare.

Preprint

Secure Computation in Decentralized Data Markets

Patent

Detecting suitability of machine learning models for datasets

Patent

Systems and Methods for Spatial Graph Convolutions with Applications to Drug Discovery and Molecular Simulation

Patent

Machine learning abstraction

2018

Journal Publication

MoleculeNet: a benchmark for molecular machine learning

Journal Publication

PotentialNet for molecular property prediction.

Journal Publication

Solving the RNA design problem with reinforcement learning

Preprint

Tokenized Data Markets

2017

Journal Publication

Low Data Drug Discovery with One-Shot Learning

Journal Publication

Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models.

Journal Publication

Is Multitask Deep Learning Practical for Pharma?

2016

Journal Publication

Computational modeling of β-secretase 1 (BACE-1) inhibitors using ligand based approaches

Preprint

Learning Protein Dynamics with Metastable Switching Systems

Patent

Conditional iteration for a non-volatile device

2015

Preprint

Massively Multitask Networks for Drug Discovery

Patent

Non-volatile key-value store

2014

Peer-Reviewed Conference Publications

Understanding Protein Dynamics with L1-Regularized Reversible Hidden Markov Models

Peer-Reviewed Conference Publications

NVMKV: A Scalable and Lightweight Flash Aware Key-Value Store

2013

Peer-Reviewed Conference Publications

The extended parameter filter

Peer-Reviewed Conference Publications

Dynamic scaled sampling for deterministic constraints