Yash Katariya
Senior Software Engineer at Google DeepMind
United States
Summary
🤩
Rockstar🎓
Top SchoolYash Katariya is a Senior Software Engineer with a decade of experience building and maintaining production-grade machine learning infrastructure and libraries, currently working at Google DeepMind. He has deep hands-on expertise in ML frameworks (TensorFlow, JAX, Flax) and compiler/runtime work—contributing to high-profile projects like TensorFlow, XLA, and JAX to improve installation, serialization, checkpointing, and hardware backend support. Beyond core engineering, Yash strengthens ecosystem reliability as a tester and technical writer, adding doctests, improving tutorials and localization, and hardening preprocessing and dataset code across Keras and TFDS. His contributions span backend C++/DevOps changes for GPU/TPU builds to Python-level model and dataset work, showing versatility across the stack. Notably, he has helped adapt many projects to new JAX primitives (jax.Array, NamedSharding), signaling an ability to future-proof large codebases for evolving hardware and APIs. He maintains an active GitHub and technical blog, reflecting a continual drive to learn and share practical ML engineering knowledge.
10 years of coding experience
5 years of employment as a software developer
MS Computer Science, MS Computer Science at North Carolina State UniversityBachelor of Engineering (B.E.) Computer Engineering, Bachelor of Engineering (B.E.) Computer Engineering at Pune Institute of Computer Technology
English, Hindi, Marathi, Marwari
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (58)
checkpoint10
preprocessing10
pytest10
artificial-neural-networks10
python10
translation10
preprocess10
data-set10
testing10
memory-management10
machine-learning10
data-serialization10
partitioning10
sharding10
flax10
Programming languages (9)
Github contributions (5)
jax-ml/jax
Jul 2021 - Jan 2023
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
Role in this project:
Back-end Developer
Back-end DeveloperContributions:57 releases, 1239 reviews, 609 commits in 1 year 6 months
Contributions summary:Yash's contributions primarily focused on enhancing the array serialization library within the JAX project. They implemented functionality for checkpointing, including handling of various data types like int4 and bfloat16 and contributed to memory management optimization. The user also worked on extending the functionalities of device_put and jax.jit to support explicit layout specifications and handling memory kinds, enhancing the project's capabilities in data placement and memory management.
pytorchpythonjitautomatic-differentiationgpu
tensorflow/probability
Jun 2019 - Oct 2022
Probabilistic reasoning and statistical analysis in TensorFlow
Role in this project:ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:12 commits in 3 years 4 months
Contributions summary:Yash contributed to the development of documentation tools, specifically for generating API documentation for the TensorFlow Probability library. They also added error handling in a JAX notebook, ensuring end-to-end execution. Furthermore, the user made changes to use public JAX types, indicating a focus on code maintenance and library updates.
statisticspythonprobabilistic-reasoningdata-sciencedeep-learning
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.