and I'm a

data scienti

This visualization project asks the question "Are the notoriously expensive cities in Germany, really the most expensive ones?". The analysis aligns relative purchasing power and cost of living indicators with scraped rent data (all in Python) to create an interactive map in R's leaflet package

Toy example of the common card game "Mau Mau" in vanilla Python. The game includes simple computer logic and allows for Computer vs. Computer and Human vs. Computer modes of varying difficulties.

Why not build your CV as a website? The entire project is coded in vanilla HTML5 and CSS3 with some minor functions in javascript. The website allows for a direct "print to pdf" button, which saves a pdf of the selected theme as a one-pager.

A business website for entering simple multiple choice questions for a tutoring purpose - real application for a customer in Spain. Use the login credentials of email: 'test' password: 'test' to play around with it. The website leverages a Firestroe backend connected to an Angular Frontend. It connects state persistence via cookies and dynamic styles to allow for a direct student feedback.

The very website you are looking at right now! The project leverages React.js's file patterns and component modularization as well as community packages for slider motions, typing effects, custom alerts, and smooth scrolling. Combined with Tailwind CSS and email.js as backend this results in a small but fullstack project. Fun!

This web application comes from a German-based NGO, who set out to help Corona-plagued students with homeschooling. I helped with dashboard functionalities for its tutors, indicating progress, achievements, etc. The entire app is built in Angular and finds a native translation via Ionic. My main responsibility lies in frontend visualizations (Angular + CSS3) of data.

This website represents an online class project of a React App, which fetches meal data from a Firebase backend. It leverages custom hooks, asynchronous programming, and basic state management via Context API in React.

This project aims to mimic a basic interaction with an interactive data fetching service. Github provides one of the largest project interaction systems of the world. It hosts numerous widely used repositories which we will explore in this simple application.The main purpose of the project is tos illustrate the benefits but also the limitations of the open data API and how to potentially overcome them for your own project. Have fun!

A dummy weather App, fetching weather data from a public weather API, storing marked cities in a firebase backend and presenting the frontend built off an Angular design style. The styling utilizes Angular Material for custom input fields and an extensive data table with sorting, filtering, and pagination functionality.

A fake Landing Page of a hypothetical skate shop built in vanilla HTML5 with CSS3. The focus of this project was to practice the placement without CSS techniques as Grid and Flexbox paired with some nice CSS animations.

This website mainly aims to practice basic backend (node.js) logic for user authentication, routing, and blog posting. HTML and CSS was used for basic styling, which remained a secondary concern in this project. The page represents a memorial app to post memories of the belated Kobe Bryant, who died in 2020.

The frontend implementation of my Master Thesis to visualize soccer formations. The user can change the visualization granularity over multiple tabs, integrating custom-built visualizations in D3.js. The overall frontend layout was achieved in an Angular App with organic typescript usage.

The backend portion of my Master's thesis. It handles the data fetching, cleaning and processingf steps in Python. The database for the large datasets of geospatial data is loaded from a Postgres SQL Database. More details about the implemented calculation logic can be found in the accompanying github repository.

A class project which compares different neural network architectures in classification accuracy for single sentence classification tasks. Especially with small input data, the data hungry architectures require a lot of fine-tuning to retrieve the required features to make reasonable assumptions. The classifiers are implemented in Python utilzing Keras or Tensorflow libraries.

This visualization project asks the question "Are the notoriously expensive cities in Germany, really the most expensive ones?". The analysis aligns relative purchasing power and cost of living indicators with scraped rent data (all in Python) to create an interactive map in R's leaflet package

Toy example of the common card game "Mau Mau" in vanilla Python. The game includes simple computer logic and allows for Computer vs. Computer and Human vs. Computer modes of varying difficulties.

A website to present an interaction interface for a product management system based off Google's Product Taxonomy. This (exclusively desktop styled) website allows for authorization, user interaction with the data locally, as well as storing the updated data online in a Firestore Cloud Backend. Use the user credentials 'Marc' with password 'SICK' to get started.

Why not build your CV as a website? The entire project is coded in vanilla HTML5 and CSS3 with some minor functions in javascript. The website allows for a direct "print to pdf" button, which saves a pdf of the selected theme as a one-pager.