Key milestones across Tech and Creativity are documented below.
Entries may be expanded into dedicated posts using inner-page.html as a template (e.g., methods,
results, and implementation notes).
2006–2015: Foundations (Creative + Technical)
- 2006–2008 — Audio Engineering Diploma (trained audio engineer), SAE Institute Cologne Creativity
- 2008–2010 — B.A. Recording Arts (Middlesex University London) Creativity
- 2011–2015 — M.A. Systematic Musicology (University of Cologne); thesis: C++ VST audio effect (Steinberg VST SDK) Creativity
2018–2019: Machine Learning & Public Recognition
- 2018 — Neural Machine Translation practicum: attention seq2seq RNN with beam search and n-best rescoring; BLEU improvement on Multi30k dev Tech
- 2019 — Hackathon “Smarter City – Aachen besser gestalten” (digitalCHURCH): Best Pitch (BOTIAS); public documentation available via Aachener Zeitung Tech
2020–2026: RWTH PEM (Tech in Production)
- 2020–2026 — AI-based development & IT services: privacy-first on-prem RAG, Offer/Tender generators, PowerPoint VSTO/COM add-in, enterprise IT operations; multi-site (Avantis + Battery Innovation Cluster (BIC)) Tech
- 2020–2026 — Content sources and academic program pages used for automated tender generation were maintained on official PEM pages (see: pem.rwth-aachen.de/go/id/fecr) Tech
2022–2023: XR Learning Technologies (RWTH LearnTech) + VR Analytics
- 04/2022–10/2023 — Contributions were delivered in the RWTH LearnTech environment (LearnTech), including the Unity-based project xTeach VR: Google Resonance Audio was integrated and parameterized for realistic distance and room perception; experimental stimuli were created and optimized for controlled studies. Tech Creativity
- 2022–2023 — Learning analytics instrumentation was operationalized via xAPI; keyword recognition and interaction events were transmitted from Unity to a Learning Record Store (LRS). Data was processed and visualized through a React-based dashboard to support structured analysis of learner behavior. Tech
2025: M.Sc. RWTH + Speech & Emotion Recognition
- 2025 — Speech & Emotion Recognition (ASR/SER): two models were trained (ASR and SER) via wav2vec2 fine-tuning; batch and real-time inference were supported; evaluation was conducted via WER/UAR; training was executed on RWTH HPC (SLURM). Tech
For a deeper technical discussion, the Case Study page may be used, or contact may be initiated via Contact.