Karmel Allison
Principal Engineer at Google DeepMind
Santa Monica, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKarmel Allison is a Principal Engineer based in Santa Monica with 15 years building production ML systems, currently shaping research and deployment at Google DeepMind. She blends deep academic training in bioinformatics (UCSD PhD work) and a BA in Linguistics from Berkeley with hands-on leadership across TensorFlow, robotics infrastructure, and AI strategy for major tech firms. A recurring open-source contributor to TensorFlow—improving MNIST examples with multi-GPU support, preprocessing optimizations, and SavedModel exports—she focuses on performance, reproducibility, and production readiness. Entrepreneurial and editorial, she founded CuratedAI, launched AI-focused startups, and co-owns Arcaio, demonstrating an appetite for productizing research. Her public writing and talks span science communication, policy, and creative AI (teaching neural nets to write poetry), reflecting a habit of making complex ideas accessible. Notably, she has advised CTOs on high-stakes AI partnerships, combining technical depth with strategic influence.
15 years of coding experience
13 years of employment as a software developer
Bachelor of Arts (B.A.) Linguistics, Bachelor of Arts (B.A.) Linguistics at University of California, BerkeleyUniversity of California, San Diego
Github Skills (9)
machine-learning10
tensorflow10
multi-gpu10
multiple-gpu10
exporter9
python9
data-export9
exports9
image-preprocessing8
Programming languages (6)
Github contributions (5)
tensorflow/models
Jan 2018 - May 2018
Models and examples built with TensorFlow
Role in this project:
ML Engineer
ML EngineerContributions:4 releases, 35 commits, 106 PRs in 4 months
Contributions summary:Karmel primarily focused on enhancing the MNIST example within the TensorFlow models repository. Their contributions include implementing multi-GPU support, optimizing preprocessing steps, and adding SavedModel export capabilities. The user's work also involved refactoring code, updating dependencies, and modifying testing procedures to accommodate these changes, indicating a focus on performance and model deployment.
deep-learningtensorflow
karmel/vespucci
Dec 2012 - Dec 2014
Contributions:511 commits in 2 years
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.