This is a full-stack web application built on top of an existing cash register. The customer portal enables customers to register upon their first visit via a QR code. Gym owners have their own login and can manage customers, obtaining detailed statistics and evaluations of sold tickets and products. The backend is developed with Laravel, while the frontend utilizes Nuxt.js. The application is hosted on a Hetzner server, leveraging a MySQL database, and is tested with Cypress. Furthermore, the API is documented with Swagger, and the application deployment is managed with Laravel Forge via a Docker container. This project was co-developed by my co-founder and me over the course of eight months and is currently in use at more than 200 gyms. Below, you can see a demo in a more compact version based on the original project. Username: demo@user.de, Password: linkedin.

This frontend application displays data for the logged-in car-sharing company through user-friendly and interactive graphs and maps. Built with Nuxt.js, it incorporates the Leaflet library for map visualization. Data is pulled via a RESTful API from the preexisting backend. This project was co-developed by my co-founder and me over six months and is currently deployed across more than 30 car-sharing companies.

Münster GPT was developed during the MSHACK23 and is now part of the Solution Enabler Program. This ongoing project, a collaborative effort by my team and me, connects to OpenAI's GPT-4 API and is developed in Python using Django. It utilizes a Chroma DB to store data scraped from the publicly available city website. Our project was awarded 2nd place at the hackathon. A test version of the chatbot is expected to be available by June 2023. We invite you to visit the website to learn more about ours and other projects.

The Bikecounter is an R-Shiny application that presents data through user-friendly and interactive graphs and maps. I developed this mini-project for fun over a weekend. The application, built with R-Shiny, is hosted on Google Cloud Run using a Docker container. To my knowledge, the application is still used in university lectures as an exemplary demonstration of the straightforward utilization of open data.

During my university studies and my time at the DFKI (German Research Center for Artificial Intelligence), I engaged in multiple data science projects. These projects included the cleaning and analysis of large datasets to identify correlations, trends, and outliers. Several projects also involved developing forecasting models. The work was primarily conducted using R and Python, incorporating R-Shiny for result visualization. Below, I have listed a few university projects for review, and one DFKI publication.

This Website was built using Nuxt, hosted on a Hetzner server and is meticulously optimized for both SEO and SEA, ensuring high visibility and performance. My co-founder and I collaboratively developed the website, continuously updating it with new features and content to keep it fresh and engaging.