Projects

This showcases some of the projects I have worked on in recent years. If you have any questions or want to learn more, please reach out.

Currently, I am working on FLIZAR, an AI-powered exam prepration tool for students and lectureres. Also, I participate in and am always on the lookout for Hackathons (and other events). If you can recommand any hackathon or event, please let me know.

FLIZAR Link to heading

FLIZAR originated with the Digital Seeds ideathon. Our project group formed in early-mid 2023 and since then we have worked hard on our platform.

FLIZAR enables students to prepare as effficient as possible for their exams (personalized to their specific course considering university and lecturer). This is possible by custom generating an unlimited amount of training materials for each student. The exercises include solutions and FLIZAR adapts to a learner’s performance. For lecuturers, FLIZAR provides detailed insights into their student’s understanding of the lecture material. Also, we provide exam task suggestions tailored to the specific course content and difficulty.

Currently, we test FLIZAR internationally at universities and lower education. We cooperate with professors, other educators and school book publishers. Some content is also already public and you can take a closer look at https://www.flizar.com. If you like looking at ‘invisible’ private repos, you can also look to see nothing on GitHub.

Hackathons Link to heading

Hackathons are great fun and I really enjoy getting a software ready in just a few hours or days. Some hackathons I have participated in so far are those:

  • Winner @msgSystems, Karlsruhe (Germany), GitHub Repo
  • @Optiver, Zurich (Switzerland)
  • @KFW, Frankfurth (Germany), GitHub Repo
  • @ETHZurich/@VISCon, Zurich (Switzerland), GitHub Repo
  • @EPFL/@insomni’hack, Lausanne (Switzerland)

Field Robot Development Link to heading

With a team of varying size (2 to 5), we have worked on the Field Robot Development Project. I’ve been part of the project from 2019 to 2023.

The goal was to build a small, mobile agricultural robot for row crops such as corn. The robot ought to be fully autonomous (meaning no GPS, no cellular or other connectivity etc.). Our robot drive system based on a recycled hoverboard, whose motors we reused. For the majority of the time we also worked on creating a two-weheled self-balancing system for increased mobility and flexibility. The ‘brain’ of the robot fully relied on cameras. We trained a custom convolutional neural net to detect drivable areas and used reverse image-projection to get a 3D map of the drivable environment. This custom software was implemented in ROS2. We could then use the ROS2 framework to enable odometry, mapping, navigation, and more.

The idea for the projet came from the Field Robot Compeition. I was also awarded 5th place in the national German Youth Reearch Compeition (Jugend forscht) as well as the Konrad-Zuse Youth Award for Computer Science from the Eduard-Rhein Foundation.

A collection of code of various parts of the project can be found on GitHub.

FloRaWAN Link to heading

To be completed later.

Interactive Video Link to heading

To be completed later.

Project result available here.