This is the homepage of Dr. Birte Kristina Friesel dedicated to paid work. See ~derf for non-work entries.

About Me

I am well-versed in embedded development, (interpretable) machine learning, product line engineering, research including documentation / publication / presentation of results, prototype and product development, and a bit of project management. In my spare time, I enjoy building useful hardware and software applications, ranging from low-level AVR Assembler and C++/Lua drivers for embedded peripherals to full-stack web applications for public transit users and commuters.

I studied computer science at TU Dortmund with a minor in electrical engineering; my master's thesis covered automating energy model generation for embedded peripherals. Following up on this, I researched the intersection between energy models for cyber-physical systems and performance models for software product lines at Osnabrück University, culminating in my dissertation on Performance Models for Embedded Software Product Lines. As of March 2025, I am working as a postdoctoral researcher at Osnabrück University.

My research topics include automated measurements with cheap off-the-shelf hardware, a novel interpretable machine learning algorithm specifically tailored towards energy models of embedded peripherals, and a toolchain for performance-aware configuration of kconfig-based software product lines. All of these have a strong focus on automation: users and system designers should be freed from as many tedious and repetitive manual tasks as possible. You can find details in the publication list and on my homepage at the Embedded Software Systems Group.

During my time at Osnabrück University, I have been involved in a project for resource-efficient AI in agricultural machinery from the initial proposal and funding application to the completion of work packages within the project, and a commercial research project with a manufacturer thereof. I have designed a programming course from the ground up and taught it to students, managed lecture exercises, supervised Bachelor's theses, and supervised student assistants developing a React-based web application.

I am currently focusing on modelling runtime performance attributes of software and hardware product lines as well as the intersection between product line engineering, interpretable machine learning, and disruptive memory technologies. My Institute Homepage has more information about that.

Projects

My private open-source projects are available on GitHub. The projects page lists a few samples. Scientific code and paper artifacts are available on the ESS GitLab.

Contact

You can reach me by E-Mail at b​f@fina​lr​ewind.org. If desired, you can use PGP key 19F13FDA 072CA45E 4E8EDEE0 B63118F7 196EA660 for encryption.