Anand Avati
Lecturer at Stanford University
Los Altos, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAnand Avati is a research scientist and lecturer with 26 years of software and ML experience, currently teaching Stanford's flagship CS229 and formerly researching at Apple on special projects. He is the lead author and inventor of NGBoost, the first gradient boosting method with built-in probabilistic predictions, and has deep open-source pedigree as a founding engineer and lead architect of GlusterFS. His career bridges production-grade distributed systems (Red Hat, Gluster) and cutting-edge ML research (Stanford, H2O, Apache Mahout), with notable contributions to survival analysis and scalable matrix/vector operations. Anand combines academic rigor from a Stanford PhD with hands-on engineering that spans core platform development to probabilistic ML, and is based in Los Altos, California.
26 years of coding experience
16 years of employment as a software developer
Bachelor's Degree, Computer Science and Engineering, Bachelor's Degree, Computer Science and Engineering at Visvesvaraya Technological UniversityHigh School, High School at St. Joseph's Central SchoolDoctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at Stanford University
Github Skills (31)
pytorch10
mahout10
matrix10
python10
survival-analysis10
machine-learning10
data-serialization10
data-structure10
java10
gradient-boosting10
distributed-computing10
serialization10
vector10
standard-library10
javas10
Programming languages (7)
Github contributions (5)
stanfordmlgroup/ngboost
Jun 2018 - Aug 2020
Natural Gradient Boosting for Probabilistic Prediction
Role in this project:
ML Engineer / Data Scientist
ML Engineer / Data ScientistContributions:69 commits, 12 PRs, 41 pushes in 2 years 1 month
Contributions summary:Anand contributed to the development of a survival analysis model using natural gradient boosting. Their work included implementing a `SurvBoost` class with core methods for model fitting, prediction, and sampling. They integrated survival analysis scoring rules and base models for the boosted ensemble, performing several iterations of improvements, including line search, and natural gradient optimization. They also addressed stability issues, implemented CRPS, and implemented code to select random subsets of features and examples per base-learner.
pythonpredictionnaive-bayesnatural-gradientsmachine-learning
h2oai/h2o-3
May 2014 - Jul 2014
H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Means, PCA, Generalized Additive Models (GAM), RuleFit, Support Vector Machine (SVM), Stacked Ensembles, Automatic Machine Learning (AutoML), etc.
Role in this project:Back-end Developer
Back-end DeveloperContributions:12 commits in 2 months
Contributions summary:Anand's contributions primarily involve modifications to the H2O-3 core, focusing on data structures and methods related to vector operations and frame manipulation. They added a new method for deep cloning a Frame, implemented string-based chunks, and introduced a method for joining existing clouds. The user also made changes to enhance code reusability and efficiency, like normalizing format strings. Their work includes improvements to serialization and deserialization processes, showcasing a deep understanding of the H2O platform.
xgboostgampythonk-meansautoencoders
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.