Magnus Hyttsten
Pioneering AI Quality And Security at Stay tuned for more details
Stockholm, Sweden
Summary
🤩
Rockstar🎓
Top SchoolMagnus Hyttsten is a seasoned engineering leader and founder with over two decades of experience pioneering AI, ML compilers, and high-scale data systems, currently focused on AI quality, security, and EU AI Act readiness. At Google since 2013 he helped build TensorFlow, led compiler and runtime efforts across GPUs/TPUs/XPUs, and most recently directed engineering for foundational LLM training, deployment, and generative AI evaluation for chat and coding applications. He scaled DigitalRoute from startup to a global enterprise platform handling billions of records per day and guided it to a successful exit, bringing rare operator experience across product, sales, and strategy. Magnus blends deep low-level expertise (compiler kernels to distributed model deployment) with a founder’s strategic mindset and a practical emphasis on safety, guard-rail evaluation, and real-world readiness. Based in Stockholm, he also contributes to TensorFlow examples on GitHub and is known for translating research-grade capabilities into robust, production-ready systems.
10 years of coding experience
17 years of employment as a software developer
Elementary Law, Elementary Law at Lund UniversityMathematics and Computer Science, Mathematics and Computer Science at Uppsala University
English, Spanish, Swedish, German
Stackoverflow
19reputation
2kreached
1answer
0questions
Github Skills (6)
jupyter-notebook10
swift10
tensorflow7
machine-learning6
scikit-learn6
data-science6
Programming languages (3)
Github contributions (5)
tensorflow/examples
Nov 2019 - Dec 2019
TensorFlow examples
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:17 commits, 2 PRs in 13 days
Contributions summary:Magnus's contributions center around developing and documenting a Swift-based grid maze environment. They created interactive notebooks using Colaboratory and updated existing documentation with code examples and explanations for the maze environment. Their work focused on implementing the environment's core features, including cell types, jump probabilities, and printing the maze representation.
tensorflow
mhyttsten/QARocker-app
Mar 2017 - Feb 2021
Contributions:21 pushes, 1 branch in 3 years 11 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.