David Svantesson is a Staff Software Engineer based in Stevenage with a strong background in backend development, low-level programming, and machine learning infrastructure. Over roughly six years in ML and systems roles at Arm, Graphcore, and mindtrace.ai, he has moved between languages and frameworks rapidly, leveraging deep algorithmic and mathematical knowledge to optimize performance-critical code. He contributed to the high-profile TensorFlow project by adding AArch64 JIT block reorder support and ACL adaptations, demonstrating hands-on expertise in hardware-aware optimizations. Comfortable in C, assembler, and Python, he blends systems-level thinking with cloud and distributed tooling experience from earlier roles. Known for quick learning and mentorship, he thrives on translating complex research into production-grade APIs and kernels.
3 years of coding experience
5 years of employment as a software developer
Computer Science and Engineering, Computer Science and Engineering at KTH Royal Institute of Technology
An Open Source Machine Learning Framework for Everyone
Role in this project:
Backend Developer
Contributions:6 reviews, 4 PRs, 18 comments in 7 months
Contributions summary:David contributed to the TensorFlow repository by implementing support for jit-ed block reorder on AArch64 architecture. This involved modifying several files related to the AArch64 architecture, including the JIT generator and reorder kernels, as well as adapting the reorder utilities. The commits suggest optimization efforts for specific hardware, enhancing the performance of the machine learning framework. The changes include updates and fixes to support the ACL (Arm Compute Library).
An Open Source Machine Learning Framework for Everyone
Contributions:31 pushes, 10 branches in 8 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.