Jannik Lindemann
Jannik Lindemann
[ MY JOURNEY ]
- January 2024 - Present
Meet Fabrik
At KreisKlang, we developed a digital synthesizer for exploratory sound design. With the two brilliant co-founders Erik and Miguel, we worked since January 2024 to transform a rough prototype to a finished product. Curious? You can check it out on our website.
As a CTO of KreisKlang, I led the full-stack development of Fabrik, which involved building highly technical solutions for low-latency data processing, improving UX cross-platform with 400+ beta testers, managing stakeholders, and dealing with legal requirements.
At KreisKlang, I gained hands-on experience building a JIT compiler using LLVM, doing CI/CD for our Cloud Features, and diving deep into digital signal processing.
- August 2021 - March 2025
Life at University
After completing my Dual Studies with a B.Sc. in Computer Science, I took three semesters of Mathematics at Technische Universität Berlin to deepen my theoretical knowledge, which profoundly changed my approach to problems.
Continuing with my M.Sc. in Computer Science, I focused on what I enjoyed most: Data Engineering and Machine Learning, with a focus on making such systems efficient on a large scale.
I chose to do my Master's Thesis at the DAMS Research Group on the topic of discovering and generating algebraic rewrites for ML Systems. I had the chance to dive deep into ML compilers and performance optimization under the supervision of Prof. Dr.-Ing. Matthias Böhm.
I completed my studies with a perfect grade of 1.0.
- September 2018 - July 2021
Dual Studies at Wacker Chemie AG
At Wacker Chemie AG and Duale Hochschule Baden-Württemberg, I pursued my Bachelor's Degree in a dual study program. There, I had the opportunity to work as a software engineer and data scientist in different departments. While doing an exploratory analysis of chemical recipe data from labs for my Bachelor's Thesis, I discovered my passion for data engineering and data science.
I graduated as the best student with the best thesis of the Computer Science class 2021.
[ PROJECTS ]
Improving Performance in ML Systems
I made various contributions to the open-source ML System Apache SystemDS, a system to abstract the end-to-end datascience lifecycle.
My major contribution was to design a framework for automatic rewrite discovery and integration. This framework helped to discover numerous beneficial algebraic rewrites, one simple example being 0+A → A.
Further, I identified 4 critical production bugs, implemented a sparse matrix format (DCSR), and integrated data compression across federated workers.
Manual Influencer
At Google BLISSathon 2025, we built an end-to-end pipeline for automatic short form content generation using Google's Vertex AI platform. We had a lot of fun seeing what is already possible with AI and were awarded among the Top 3 teams of the Hackathon.
It is a lot of fun to look at the entertaining results, so check it out!
Jtex - LaTeX Transpiler
Tired of repeatedly writing complex equations in LaTeX? For this reason, I have created Jtex, an easy to use superset language that simplifies exactly that process. For example$\frac{\overline{x}}{y}$becomes -- (x^-)/y;making your life easier if frequently using such formulas. You can also embed JavaScript code for generating content, which may be handy when working with changing datasets.
QML - Quantum Machine Learning Framework
Can quantum circuits learn? During my seminar at Technische Universität Berlin, I had the chance to explore this question. I built a Proof of Concept Framework for training arbitrary Quantum Circuits on (very small) datasets using Amplitude Amplification. It was a great learning experience to see what is and what isn't possible with quantum computing, because your intuition can sometimes be misleading.
So can quantum circuits learn? The answer is yes. While our general-purpose approach is highly impractical for real-world use-cases, it was a great learning experience. You can find my framework on GitHub.