Ekaterina Gonina
Freelance Software Engineer at Worktrace AI
United States
Summary
🤩
Rockstar🎓
Top SchoolEkaterina Gonina is a freelance software engineer with 14 years of expertise designing and optimizing large-scale ML infrastructure, distributed systems, and training platforms, with senior roles at Google DeepMind and Twitter. She holds a PhD in Computer Science from UC Berkeley’s Parallel Computing Lab and has deep hands-on experience with distributed RL (TensorFlow, TF-Agents), GPUs/TPUs, and performance optimization for synchronous training. Her work spans hyperparameter tuning, model benchmarking and profiling, and end-to-end data pipelines (batch and streaming), including time-series metrics systems. At Google she contributed to TF-Agents and built tuning mechanisms for large-scale TensorFlow models; her open-source refactor of replay buffers in a widely used RL library highlights her focus on robust, reusable infrastructure. After a planned two-year caregiving and learning break, she’s returned to freelance work focused on AI infrastructure intersecting with interactive tooling and data visualization. She combines research-grade performance engineering with pragmatic production experience, making her especially effective at turning ML research into scalable, maintainable systems.
14 years of coding experience
12 years of employment as a software developer
MS/PhD Computer Science, MS/PhD Computer Science at University of California, BerkeleyBachelor of Science (B.S.) Computer Science, Bachelor of Science (B.S.) Computer Science at University of Illinois Urbana-Champaign
english, russian
Github Skills (7)
machine-learning10
agent10
tensorflow10
ai-agent10
python10
reinforcement-learning10
testing9
Programming languages (6)
Github contributions (5)
tensorflow/agents
Nov 2018 - Dec 2022
TF-Agents: A reliable, scalable and easy to use TensorFlow library for Contextual Bandits and Reinforcement Learning.
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:3 reviews, 49 commits, 1 PR in 4 years 1 month
Contributions summary:Ekaterina focused on refactoring and modifying the `BatchedReplayBuffer` class and its associated tests, moving it to subclass the `ReplayBuffer` base class and altering its API. They also renamed the module and class to `tf_uniform_replay_buffer` and `TFUniformReplayBuffer`. Further changes included the correction of a typo, and modifications to the DqnAgent and its related tests. Finally, they made updates in response to the change in the `tf_environment` API. These changes suggest a focus on improving the replay buffer's functionality within the reinforcement learning framework and overall codebase refactoring.
scalabletf-agentsmultiagent-reinforcement-learningtensorflow-librarymulti-armed-bandits
egonina/pycasp
Feb 2013 - Dec 2013
Contributions:28 commits, 2 comments in 10 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.