Portfolio

Featured Projects

Products and platforms I designed, built and run

flatmo.app – CMS & Accommodation Booking
Ongoing engagement

flatmo.app – CMS & Accommodation Booking

Current client work on flatmo.app: building a complete content management system and an accommodation booking application in the style of Airbnb and Booking.com.

Key Highlights

  • Complete CMS development
  • Accommodation booking application
  • End-to-end product development
Debevet UG – Backend Development
Past engagement

Debevet UG – Backend Development

Client project for Debevet UG, which runs the debevet.de veterinary practice software, the around.pet pet health app and the Wolf & Tiger veterinary practices. My contribution was building the application backend, with responsibility focused on server-side development.

Key Highlights

  • Application backend development
  • Server-side implementation for debevet.de and around.pet
around.pet
Past engagement

around.pet

Pet platform for around.pet GmbH. I built the container-native MVP with a full CI/CD pipeline, then later maintained the existing Flutter app instead of replacing it.

Key Highlights

  • Container-native MVP with CI/CD
  • Later maintenance of the existing Flutter app

Tech Stack

  • Flutter
  • Node.js
  • TypeScript
  • PostgreSQL
  • Docker
  • CI/CD
Auftrag One
Live2022 – Present

Auftrag One – AI Procurement Platform

AI platform for German public procurement (Vergabe). Aggregates tender notices from 200+ German, Austrian and EU sources, classifies each into 331 topics across 25 sectors, and scores them against a company profile.

Key Highlights

  • Event-driven micro- & macroservice architecture (Kafka Pub/Sub)
  • 530k+ notices analysed
  • ~8.5k new every week
  • AI topic classification and relevance scoring per company profile
  • Automated document analysis with deadline
  • eligibility and award-criteria extraction
  • Vergabe network graph of buyers and bidders
  • Multi-organization role-based access and team management
  • Deployed on GKE with CI/CD
  • auto-scaling and automated GCP & Backblaze backups

Tech Stack

  • Kotlin
  • Spring Boot
  • Golang
  • Apache Kafka
  • Elasticsearch
  • MongoDB
  • GCP Storage
  • Gemini AI
  • Kubernetes (GKE)
  • GitHub Actions
OnSand
Live2025 – Present

OnSand – Beach Volleyball Matchmaking

Location-based matchmaking app for beach volleyball. Players find open games and skill-rated partners nearby, organize matches in one place instead of scattered group chats, and climb city and club rankings.

Key Highlights

  • Location-based game discovery filtered by skill level
  • Automatic skill rating from Rookie to Pro with streaks
  • Match formats for 2v2
  • 3v3
  • 4v4 and King of the Court
  • Scheduling
  • team chat and cost splitting in one flow
  • Self-updating city and club leaderboards

Tech Stack

  • Flutter
  • Dart
  • Golang
  • Rust
  • PostgreSQL
  • Meilisearch
  • Kubernetes (GKE)
  • GCP
OnSand Cloud
Live2025 – Present

OnSand Cloud – Club Management

Club management software for beach volleyball associations: members, events and communication in one calm workspace, connected to the public tournament side of OnSand.

Key Highlights

  • Member directory with roles
  • status and history
  • Event scheduling with confirmations
  • waitlists and reminders
  • Direct messages
  • group chats
  • announcements and push notifications
  • Role-based access with EU-only data processing
  • CSV export for full data portability
  • Free tier for clubs up to 15 members

Tech Stack

  • TypeScript
  • Next.js
  • Golang
  • Rust
  • PostgreSQL
  • Meilisearch
  • Kubernetes (GKE)
  • GCP
Artconnect
LiveBuilt pre-AI2020-2022

Artconnect Re-Architecture

Complete system modernization from Ruby monolith to Kotlin microservices. Achieved 99.95% uptime with 60% infrastructure cost savings.

Key Highlights

  • Monolith-to-microservices migration
  • Zero-downtime deployment
  • CQRS pattern implementation
  • 60% cost reduction

Tech Stack

  • Kotlin
  • Spring Boot
  • Kafka
  • Elasticsearch
  • Kubernetes
  • Helm
  • GCP
LiveBuilt pre-AI2016-2018

High-Performance Web Crawler

Enterprise-grade web crawler processing >150k documents/day with intelligent data extraction and commercialization platform.

Key Highlights

  • >150k documents/day processing
  • Intelligent data extraction
  • Commercialization platform
  • High availability design

Tech Stack

  • Java
  • Kotlin
  • PHP
  • JavaScript
  • Distributed Systems
Private repository
Gesetze.ai – Semantic Law Search
WIP2024 – Present

Gesetze.ai – Semantic Law Search

Experimental side project for semantic search in German legal texts. Still under construction.

Key Highlights

  • Kotlin-based web crawler for legal corpora
  • Golang backend for search indexing
  • Semantic search pipeline using embedding vectors

Tech Stack

  • Golang
  • Kotlin
  • Elasticsearch
  • OpenAI Embeddings
Visit WebsitePrivate repository
Live2024

d8.gg – URL Shortener

A minimalistic, self-hostable URL shortener written entirely in Go.

Key Highlights

  • 100% Go codebase
  • Lightweight and fast
  • Designed for self-hosting

Tech Stack

  • Go
  • SQLite
  • Docker
Live2024

GitHub Actions Auto-Scaler

Personal Dockerized GitHub Actions runner that dynamically registers runners on demand. Bypasses frequent token rotation issues.

Key Highlights

  • Self-hosted CI scalability
  • Solves ephemeral token refresh problem
  • Suitable for personal GHA scaling

Tech Stack

  • Docker
  • Bash
  • GitHub Actions API
Completed2023

ImmoScout Realtime Crawler

Developed for a friend's master's thesis: a Kotlin-based crawler for ImmoScout24 that tracks listings in near real-time.

Key Highlights

  • Near real-time updates
  • Resilient scraping & filtering logic
  • Designed to track listing activity for academic analysis

Tech Stack

  • Kotlin
  • Playwright
  • Jsoup
  • HTML Parsing
Live2024

6bit.ai – Website to Markdown Converter

A web service that intelligently converts website content to clean, structured markdown format, preserving formatting and structure while removing clutter.

Key Highlights

  • Intelligent content extraction
  • Clean markdown output
  • Preserves document structure
  • Removes ads and navigation clutter

Tech Stack

  • Web Scraping
  • HTML Parsing
  • Natural Language Processing
  • Markdown Generation
Visit WebsitePrivate repository
Completed2024

Northdata Company Data Crawler

Automated web crawler built with TypeScript and Playwright to extract company information from Northdata. Handles dynamic content loading and produces structured data.

Key Highlights

  • Automated company data extraction
  • Handles dynamic JavaScript content
  • Robust error handling and retry logic
  • Structured data output for downstream processing

Tech Stack

  • TypeScript
  • Playwright
  • Web Automation
  • Data Extraction
Private repository
Open for new projects

Have a similar project?

Architecture, full-stack delivery, cloud and AI β€” tell me what you're building.