Fullstack Webdeveloper.

This website is a showcase of my projects, skills, and experience in software development. It features a modern design with a clean layout and intuitive navigation. The site is build with Next.js, Tailwind CSS and Framer Motion.

This project focuses on the implementation, optimization, and visualization of routing algorithms using OpenStreetMap data. The main objective is to develop an interactive web-based application that compares various routing algorithms, such as breadth-first search, Dijkstra, and A*, and visually represents their operation. A key contribution is the implementation of an efficient graph contraction technique, which significantly reduces computational effort by selectively eliminating intermediate nodes. Evaluation on different datasets (urban, rural, highway networks) demonstrates runtime improvements ranging from 10% to 95%, depending on the network structure and the chosen algorithm. For highway data, which has a low branching density, a node reduction of over 93% was achieved, leading to a computation speedup by a factor of 20. The developed application provides valuable insights into the functioning of routing algorithms and highlights the substantial optimization potential of structural graph transformations for efficient navigation in real-world road networks.

This project is a self-hosted AI assistant that enables users to interact with PDF documents by answering specific questions based on their content. Users can upload, manage, and view PDFs directly within the application. An on premise LLM retrieves the most relevant document sections using advanced similarity search, ensuring accurate responses. The system includes a built-in chat history to track past interactions and a secure authentication mechanism for controlled access. Designed for on-premise deployment, it guarantees full data privacy and independence from external AI providers. The platform features a modern Next.js frontend and a Python FastAPI backend, all deployed in a Docker container for streamlined management. It is ideal for researchers, legal professionals, businesses, and institutions that need an AI-powered document assistant to extract insights from extensive PDF collections while maintaining full control over their data.

This project is a web application for managing and approving quote numbers. Users with the appropriate permissions can create new quote numbers, while others must approve them. The system features secure authentication and supports different user roles to control the management and approval process. The application is built with Next.js and uses Supabase as its database, providing a modern, scalable, and secure solution for businesses.

A community-based Flutter application that connects neighbors to help each other with everyday tasks and needs. Nabelink is a mobile application designed to foster community connections by allowing users to both request help and offer assistance to others in their neighborhood. Whether it's moving furniture, walking a dog, or borrowing kitchen supplies, Nabelink makes it easy to connect with neighbors and build a stronger community.