Jens Elofsson
GPU SW Engineer at Arm
Sweden
Summary
🤩
Rockstar🎓
Top SchoolJens Elofsson is a GPU software engineer at Arm with 11 years of embedded and ML systems experience, specializing in drivers and firmware for Mali GPUs and optimized kernels for resource-constrained targets. He has deep C and low-level systems expertise from developing physical-layer L1 software for NB-IoT, DSP algorithms, and Linux-based embedded applications, and has applied that to TensorFlow Lite Micro and CMSIS-NN contributions. His open-source work includes implementing an SVDF kernel and fixing memory and FPU issues in widely used projects like tflite-micro and ARM's CMSIS_5, reflecting careful attention to correctness and safety in constrained environments. Based in Sweden and holding a Master’s in Information Technology from Lund University, he combines hands-on kernel- and driver-level coding with a pragmatic focus on portability and device-specific optimization. An understated strength is his track record of tracking down subtle bugs (out-of-bounds, toolchain FPU issues) that improve reliability across multiple embedded targets.
11 years of coding experience
6 years of employment as a software developer
Master's degree, Information Technology, Master's degree, Information Technology at The Faculty of Engineering at Lund University
Swedish, English
Stackoverflow
37,106reputation
1.3mreached
37answers
1question
git
top-1%
gitignore
top-1%
Github Skills (27)
git10
c-language10
architecture10
gitignore10
c1110
makefile10
n10
c1710
cmsis10
sys10
arm10
embedded10
architectures10
cprogramming-language10
build-system9
Programming languages (6)
Github contributions (5)
tensorflow/tflite-micro
Jun 2021 - Mar 2022
Infrastructure to enable deployment of ML models to low-power resource-constrained embedded targets (including microcontrollers and digital signal processors).
Role in this project:
Embedded Systems Engineer / IoT Developer
Embedded Systems Engineer / IoT DeveloperContributions:1 review, 8 commits, 8 PRs in 8 months
Contributions summary:Jens contributed to the tflite-micro repository by implementing and modifying code related to embedded systems and machine learning model deployment on resource-constrained devices. Their work included adding kernel sources and adjusting build configurations. They addressed an fpu-related bug in gcc, fixed warnings in memory management code and updated the Arm Ethos-U driver and kernel, involving configuration changes and driver updates. Additionally, the user increased stack size for a target architecture and enabled tests for a specific target, showcasing expertise in embedded systems and device-specific optimization.
signalml-modelslow-powerprocessorsdeployment
ARM-software/CMSIS_5
Aug 2020 - Nov 2021
Role in this project:Embedded Systems Engineer / IoT Developer
Embedded Systems Engineer / IoT DeveloperContributions:5 reviews, 5 commits, 2 PRs in 1 year 3 months
Contributions summary:Jens contributed to the implementation of a Signed Value Decomposition Filter (SVDF) kernel within the CMSIS-NN library. Their work involved adding the core SVDF functionality, including the `arm_svdf_s8` function, and subsequent updates to address review comments and fix potential read-out-of-bounds bugs. Further contributions addressed code warnings and applied necessary fixes to ensure code quality and safety.
cortexcortex-artoscmsismicrocontroller
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.