David Norman
Distinguished Engineer at Graphcore
Bristol, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolDavid Norman is a Distinguished Engineer with over two decades of hands-on experience designing and delivering high-performance, real-time and embedded systems across industries from broadcast media to AI accelerators. He has led teams of 4–60 engineers through roles spanning senior engineering to chief software architect at Graphcore, driving machine learning frameworks, microservices, and cloud-native delivery for high-performance hardware. Proficient in C, C++, C#, and Java across Windows, macOS and Linux, he combines low-level expertise (assemblers, traces, oscilloscopes) with large-scale system architecture and multi-core design. His open-source contributions to the XLA ML compiler highlight a focus on low-level compiler optimizations and support for emerging data types like F16, underscoring a practical bridge between ML software stacks and hardware efficiency. Known for fast comprehension and pragmatic trade-offs between performance, scalability, cost and delivery time, he thrives both as an individual contributor and a strategic technical leader. Based in Bristol with a physics degree from the University of Bristol, he brings a methodical, measured approach rooted in instrumentation and problem-solving.
10 years of coding experience
30 years of employment as a software developer
A levels Physics, A levels Physics at HSDC AltonBachelor’s Degree Physics, Bachelor’s Degree Physics at University of Bristol
Stackoverflow
73reputation
4kreached
7answers
1question
Github Skills (19)
c-language10
tensorflow10
compiler-optimization10
xla10
compiler10
cprogramming-language10
machine-learning9
datastructures-algorithms8
data-structures8
data-structure8
algorithms8
algorithm8
nvidia6
macos6
server6
Programming languages (6)
Github contributions (5)
openxla/xla
Feb 2017 - Sep 2019
A machine learning compiler for GPUs, CPUs, and ML accelerators
Role in this project:
Back-end Developer
Back-end DeveloperContributions:69 commits in 2 years 6 months
Contributions summary:David primarily contributed to the XLA compiler, focusing on low-level improvements and optimizations. Their work included converting TensorFlow operations to use elementwise XLA ops, enhancing allocation tracking for pre-tracked buffers, and adding support for F16 data types within literals. These changes demonstrate a focus on improving the efficiency and capabilities of the compiler, specifically improving the underlying infrastructure that supports ML models. The user also made changes to the XLA slice operation including the addition of a strides parameter.
compilercommunity-drivenmachine-learningmodular
DavidNorman/tensorflow
Apr 2017 - Jul 2019
Computation using data flow graphs for scalable machine learning
Contributions:132 commits, 315 pushes, 144 branches in 2 years 3 months
computationscalabledata-sciencemachine-learninggraphs
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.