Fredrik Knutsson
Principal Engineer, Machine Learning at Arm
Malmo, Skåne County, Sweden
Summary
🤩
Rockstar🎓
Top SchoolFredrik Knutsson is a Principal Engineer in Machine Learning at Arm with over a decade of experience designing and delivering embedded and system software across telecommunications, automotive and semiconductor domains. He combines hands-on low-level coding (C, assembly) and system architecture with team leadership and product planning, having led engineering groups and managed hiring and CI/CD improvements. Fredrik’s open-source contributions to projects like ARM CMSIS-NN and IREE show a practical focus on optimizing neural network primitives for constrained devices and enabling arm64 CI builds and faster Docker builds. Comfortable moving features from concept through design, implementation and test, he has deep experience in NB‑IoT/LTE, AUTOSAR automotive gateways and resource-efficient ML on ARM processors. Colleagues know him for pragmatic architecture choices that balance safety, performance and manufacturability—skills honed across Arm, Bosch, Ericsson and consultancy work.
10 years of coding experience
14 years of employment as a software developer
Free standing course Systems Programming in the UNIX/Linux Environment 7.5 credits, Free standing course Systems Programming in the UNIX/Linux Environment 7.5 credits at Mid Sweden UniversityMaster of Science Wireless communication, Master of Science Wireless communication at Chalmers University of Technology
English, Swedish
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (26)
github-ci10
docker10
c1110
n10
dockers10
cicd10
gcp10
c1710
cmsis10
sys10
build-automation10
neural-network10
arm10
embedded10
githubaction-workflow10
Programming languages (4)
Github contributions (5)
iree-org/iree
Nov 2021 - Dec 2023
A retargetable MLIR-based machine learning compiler and runtime toolkit.
Role in this project:
DevOps Engineer
DevOps EngineerContributions:44 reviews, 5 PRs, 91 comments in 2 years 1 month
Contributions summary:Fredrik's contributions primarily focused on enhancing the Continuous Integration and Continuous Delivery (CI/CD) infrastructure for the IREE project. This involved adding support for arm64 architecture in the CI/CD pipelines, including the setup for the GCP runner with gcloud CLI support. Furthermore, the user updated the Docker build process, particularly for arm64, and optimized build times with ccache. These efforts aimed to streamline and improve the build and testing processes across various architectures.
mlirspirvvulkantensorflowcompiler
ARM-software/CMSIS_5
Mar 2019 - Nov 2019
CMSIS Version 5 Development Repository
Role in this project:Embedded Systems Engineer / IoT Developer
Embedded Systems Engineer / IoT DeveloperContributions:8 reviews, 5 commits, 19 PRs in 8 months
Contributions summary:Fredrik primarily contributed to the CMSIS-NN repository, focusing on optimizing and enhancing neural network functions for embedded systems. Their work involved fixing potential overflow issues in the `arm_nnsupportfunctions.h` file, specifically related to shifting operations. The user also implemented and refined buffer size calculations and added batch processing capabilities to convolution kernels, improving the efficiency of neural network computations on resource-constrained devices. These modifications suggest a focus on improving the performance and accuracy of neural network operations on ARM processors.
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.