Marissa Ikonomidis
Staff Software Engineer at Google
Sunnyvale, California, United States
Summary
👤
Senior🎓
Top SchoolMarissa Ikonomidis is a Staff Software Engineer in Sunnyvale with a decade of experience building high-performance systems for ML serving and Android platform subsystems. As tech lead at Google, she directs a component that optimizes TensorFlow, JAX and Pathway models across accelerators—focusing on execution planning, memory reduction, partitioning, and LLM-serving performance. Her open-source contributions to TensorFlow include meaningful batching and XLA GPU fixes, showing hands-on impact on a widely used ML framework. Previously she led cross-company efforts to standardize Android buffer stacks and improved on-device power modeling, combining low-level kernel expertise with large-scale coordination. Known for rigorous CS fundamentals (4.0 Computer Science degree from Georgia Tech) and a pragmatic approach to performance engineering, she thrives at the intersection of compilers, drivers, and production ML infrastructure.
9 years of coding experience
4 years of employment as a software developer
Bachelor's degree, Computer Science, 4.0, Bachelor's degree, Computer Science, 4.0 at Georgia Institute of Technology
English
Github Skills (7)
tensorflow10
python10
cprogramming-language9
machine-learning9
c-language9
deep-learning9
mlops7
Programming languages (5)
Github contributions (5)
tensorflow/tensorflow
Jun 2020 - Jan 2023
An Open Source Machine Learning Framework for Everyone
Role in this project:
Back-end Developer
Back-end DeveloperContributions:31 commits, 2 comments, 1 issue in 2 years 7 months
Contributions summary:Marissa's contributions center around optimizing the TensorFlow framework, specifically focusing on improving the efficiency and functionality of batching operations. They addressed memory management issues in the XLA GPU compiler by correcting assumptions about int32 operations and added support for rewriting batching options during model loading. Furthermore, they refactored the batching code and added a new batching padding policy. These changes demonstrate a focus on improving the performance and flexibility of the TensorFlow's batching capabilities.
pythondata-sciencedeep-learningmlmachine-learning
Joshndroid/android_device_htc_flounder
Sep 2016 - Mar 2017
Contributions:68 commits in 6 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.