Yogesh Garg
Member Of Technical Staff at Microsoft AI
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolYogesh Garg is a Member of Technical Staff at Microsoft AI with 14 years of experience building scalable data, ML and experimentation platforms across companies like Weights & Biases, Uber, Databricks and Tesla. He combines deep systems and ML engineering expertise—designing distributed deep learning on Spark, speeding perception training pipelines, and owning platforms that run thousands of weekly A/B analyses—to deliver measurable product and performance gains. At Uber he boosted analysis success rates from 65% to 90% and designed pre-computation pipelines to cut memory footprints; at Databricks he contributed to Spark Deep Learning and a Keras-based estimator used in large-scale pipelines. Comfortable across backend, infrastructure and production ML, he also contributed substantive open-source fixes around thread-safety, parameter validation and test refactoring in the popular spark-deep-learning project. Trained at Columbia and IIT Delhi with an exchange at KTH, he blends rigorous academic foundations with pragmatic, production-first delivery.
14 years of coding experience
8 years of employment as a software developer
Indian Institute of Technology Delhi (IIT Delhi)Exchange Student Mathematics and Computer Science, Exchange Student Mathematics and Computer Science at KTH Royal Institute of TechnologyMaster’s Degree Computer Science, Master’s Degree Computer Science at Columbia University
Github Skills (10)
keras10
apache-spark10
machine-learning10
deep-learning10
pytest10
python10
image-processing9
validation9
validate9
cross-validation8
Programming languages (11)
Github contributions (5)
databricks/spark-deep-learning
Apr 2018 - Aug 2018
Deep Learning Pipelines for Apache Spark
Role in this project:
ML Engineer
ML EngineerContributions:9 commits, 16 PRs, 3 pushes in 4 months
Contributions summary:Yogesh primarily contributed to the `KerasImageFileEstimator` class, which is part of a deep learning pipeline for Apache Spark. Their work focused on improving the estimator's functionality, addressing model tuning issues, and enhancing parameter validation. The commits include modifications to the `fitMultiple` method, ensuring thread safety, and implementing better error handling, and include a substantial refactoring of the testing framework.
data-sciencedeep-learningmachine-learningapachespark
yogeshg/small-projects
Jan 2016 - Jul 2024
A repo to contain code for all completed / uncompleted small projects that do not deserve a full repository.
Contributions:6 PRs, 127 pushes, 8 branches in 8 years 7 months
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.