Ken Franko is a Senior Staff Software Engineer based in the San Francisco Bay Area with 6 years of experience building large-scale ML infrastructure and TPU-optimized systems at Google and Google DeepMind. He has led production LLM serving efforts that power internal products like Search and Gemini as well as Cloud customers, improving reliability through smarter load balancing, load shedding, and observability. His background in high-performance scientific computing (Sandia National Labs) and a Ph.D. in Aeronautics and Astronautics from Stanford gives him deep expertise in parallel performance, debugging at scale, and numerical reliability. Ken is an active contributor to TensorFlow, where he has both refactored TPU-related estimator code and improved API documentation—demonstrating a focus on developer experience as well as backend performance. He combines systems-level engineering with a knack for clarifying complex APIs, making ML infra both faster and easier to use. A less obvious strength is his history of shipping production database and backend services for consumer-facing products early in his career, showing fluency across research, infra, and product domains.
6 years of coding experience
12 years of employment as a software developer
BS/BA Aerospace Engineering and Spanish, BS/BA Aerospace Engineering and Spanish at Case Western Reserve University
Doctor of Philosophy (Ph.D.) Aeronautics and Astronautics, Doctor of Philosophy (Ph.D.) Aeronautics and Astronautics at Stanford University
An Open Source Machine Learning Framework for Everyone
Role in this project:
Back-end Developer & Documentation Specialist
Contributions:7 reviews, 289 commits, 6 PRs in 3 years 1 month
Contributions summary:Ken primarily contributed to the documentation of the TensorFlow library. They updated documentation for various APIs, including `tf.get_logger()`, `DataLossError`, `tf.AggregationMethod`, `crop_and_resize`, and `VariableAggregation`. Their contributions involved clarifying wording, adding examples, and improving the overall clarity of the documentation, demonstrating a focus on developer experience. They also removed an API, reflecting contributions beyond mere documentation.
Contributions summary:Ken primarily focused on modifying and refactoring the TensorFlow Estimator library. Their contributions involved renaming functions, moving test files, and updating existing code related to TPU (Tensor Processing Unit) functionality. The changes indicate a focus on code organization and adapting the library for specific hardware acceleration using TPUs. This includes modifications in the evaluation and export functionalities of the library.
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.