A selection of my work: from platform engineering to machine learning.
An RL project for Super Mario Bros: all 32 levels solved, the game finished end to end (Double DQN + PPO), with Go-Explore, behavior cloning and transfer learning – and five metrics that measured something other than what was meant.
Python PyTorch Double DQN PPO Stable-Baselines3 Go-Explore
Automated load forecasting with AutoTS: genetic model search, rolling-window evaluation across a full year and template-based persistence of the best models.
Python AutoTS pandas NumPy
Rolling-window backtesting pipelines for building energy load forecasting: LightGBM, NHiTS, TSMixer and a robust seasonal baseline.
Python Darts LightGBM NHiTS TSMixer pandas
Research thesis at the Fraunhofer Institute for Solar Energy Systems (ISE) on data-driven methods in energy systems.
LaTeX Python Research Energy Systems
A shop single-page application with product search, cart, internationalization and theming, built in the "Web Programming with React" course at HFU.
React JavaScript Vite REST i18n
A local Kubernetes lab following the App-of-Apps pattern, running end-to-end: ArgoCD reconciles the cluster against the Git repo – including ingress, cert-manager, monitoring and a demo app, with CI building the images to GHCR.
Kubernetes ArgoCD Helm Go kind GitHub Actions