I am a C++ engineer focused on performance: algorithm engineering, parallel and distributed computing, and GPU programming. I like working close to the hardware: profiling the hot path, choosing the data layout, and checking with measurements whether an optimization really holds up.

I completed my M.Sc. in Computer Science at the Karlsruhe Institute of Technology (KIT) in September 2026 with a grade of 1.5. I am looking for a full-time position in high-performance C++ (low latency, HPC and GPU, or performance-critical product software) and am available now.

Technical focus: C++ (C++11 to C++23, templates and metaprogramming), CUDA, MPI, Python · Linux, CMake, Git · profiling, multithreading, cache-aware data structures, parallel and distributed algorithms.

How it started

C++ was my first programming language. I taught myself in middle school from the book Grundkurs C++ by Jürgen Wolf, and it has stayed my main language ever since.

My first formal training came at the Berufskolleg Technik in Siegen (2015 – 2018), where I qualified as a state-certified information technology assistant and earned the Fachhochschulreife: programming fundamentals, computer systems and networking. Alongside school I joined the junior studies program of the University of Siegen (2016 – 2017) and attended computer science lectures there, among them Software Engineering I and Linear Algebra for Computer Scientists.

Industry

All my industry work so far ran alongside school and university.

pmdtechnologies · Siegen · 2017 – 2022

pmd develops 3D time-of-flight depth sensors. I started with an internship in the summer of 2017: an NSIS installer for one of the main products and smaller C++ tickets across the code base, with Jira, Bitbucket and Jenkins as the team’s toolchain. I stayed on for almost five years, alongside school and my bachelor’s. Most of that work was C++11 with Qt and OpenGL: I re-implemented the raw-data viewer in C++/Qt and worked on the 3D visualization. For Android I wrapped the native C++ sensor framework in a Java library (JNI) and built a visualizer on the Camera2 API to test the sensors’ Camera2 integration. This is where C++ turned from a hobby into a professional tool for me.

Vector Informatik · Karlsruhe · 2023 – 2026

At Vector I worked in the medical-devices department, on an implementation of IEEE 11073 SDC, the standard for networked communication between medical devices. My first tasks there were outside C++; over time I moved into the C++20 part of the stack. I implemented parsers for the SDC message family, built on template metaprogramming that makes the message model available at compile time and generates the parsing code from it. In this domain the constraint is certifiable correctness rather than raw speed, and it taught me to be precise about what code guarantees and what a test actually shows.

Research

Lattice Boltzmann Methods group · H-BRS · 2022 – 2024

My bachelor’s thesis led me to lettuce, an open-source, PyTorch-based framework for GPU-accelerated Lattice Boltzmann simulations of fluid flows, and I kept working on it as a student research assistant (three contracts, 16 months in total, remote from Karlsruhe). I replaced generic PyTorch operations with custom CUDA kernels and wrote a hardware-aware code generator that JIT-compiles kernels for the configuration at hand. In my thesis, the generated kernels reached up to 46× the throughput of the PyTorch baseline.

Lattice Boltzmann Research Group · KIT · 2022 – 2023

In my first year at KIT I also worked on the C++ computational engine of the Lattice Boltzmann Research Group.

Master’s thesis: KaSpan · KIT · 2025 – 2026

In Prof. Peter Sanders’ Algorithm Engineering group I built KaSpan, a distributed algorithm for strongly connected components in C++ and MPI, and evaluated it on the HoreKa supercomputer with up to 8,208 cores against the existing distributed solvers. The full story, with plots, is in the post on KaSpan; the code is on GitHub.

Education

M.Sc. Computer Science · Karlsruhe Institute of Technology · 2022 – 2026

I specialized in parallel computing and algorithm engineering. The courses I enjoyed most were the ones closest to how fast code really runs: Parallel Algorithms, Advanced Data Structures, Text Indexing and Algorithm Engineering, plus the lab on efficient parallel C++. A highlight outside that focus was the practical challenge in Formal Systems: a generator for SAT encodings, written in Java, and my submission was the best of the course. The master’s thesis on KaSpan was graded 1.3, and my overall grade is 1.5 (German scale, 1.0 = best). Because I worked throughout my studies, the degree took longer than the standard two years.

Outside the lectures I was active in the student council (student counseling and interviews for the master’s admission), in the student union’s festival committee and in the program committee of the AKK cultural center.

B.Sc. Computer Science · University of Siegen · 2018 – 2022

My specialization was visual computing. For my bachelor’s thesis, Efficient CUDA and C++ Implementation of Fluid Dynamics Simulations in a PyTorch-Based Open-Source Framework (grade 1.0), I brought CUDA and C++ into lettuce, which was the start of my research work on GPU performance.

Contact