Sabela Ramos
Staff Software Engineer at Google DeepMind
Zurich, Zurich, Netherlands
Summary
👤
Senior🎓
Top SchoolSabela Ramos is a Staff Software Engineer at Google DeepMind with a decade of experience bridging research and production in ML and systems. Her background includes a postdoc at ETH Zurich focused on performance modeling and algorithm design for shared-memory architectures, and a PhD in Computer Science from Universidade da Coruña. At DeepMind and Google she has contributed to open-source ML infrastructure—notably work on the Acme RL library and TensorFlow Datasets integrations for D4RL—showing strength in dataset engineering, policy APIs, and MLOps. She combines research rigor with production-grade engineering, refactoring core components and extending dataset builders for real-world RL workflows. Based in Zurich and active in initiatives that promote women in tech, she brings both technical leadership and community engagement to cross-disciplinary teams. An often-overlooked asset is her sustained track record of moving academic ideas into maintainable, widely used code.
10 years of coding experience
13 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at Universidade da Coruña
Spanish, Galician, english (proficiency), French, German
Github Skills (12)
machine-learning10
jax10
tensorflow10
python10
reinforcement-learning10
data-engineering9
agent9
hdf8
data-pipeline8
data-pipelines8
githubaction-workflow7
github-ci7
Programming languages (4)
Github contributions (5)
google-deepmind/acme
Sep 2021 - Sep 2022
A library of reinforcement learning components and agents
Role in this project:
ML Engineer
ML EngineerContributions:20 commits in 1 year
Contributions summary:Sabela primarily contributed to the Acme library, focusing on reinforcement learning components. Their commits include modifications to integrate `rlds` transformations within the `acme/datasets/tfds.py` file, demonstrating an understanding of data loading and transformation for reinforcement learning datasets. They also refactored and updated code related to `policy_network_factory` and `make_policy`, contributing to the agent build process and infrastructure. Furthermore, the user updated the codebase to utilize the `make_policy` API for different baseline implementations.
reinforcement-learningreinforcementagentsdeep-reinforcement-learning
tensorflow/datasets
Dec 2020 - Jul 2022
TFDS is a collection of datasets ready to use with TensorFlow, Jax, ...
Role in this project:Back-end Developer & MLOps Engineer
Back-end Developer & MLOps EngineerContributions:78 commits in 1 year 7 months
Contributions summary:Sabela's commits primarily focus on enhancing the `tensorflow/datasets` repository with new datasets and features related to reinforcement learning, particularly in the context of the RL Unplugged project. They added the D4RL HalfCheetah, Hopper, and Walker2d datasets and incorporated support for dataset metadata. Furthermore, the user refactored the D4RL builder for greater flexibility, enabling support for Adroit datasets. The user demonstrated expertise in integrating and managing datasets within the TensorFlow ecosystem.
datanumpydeep-learningdatasetmachine-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.