Applied AI Researcher.
Engineer. Builder.
I build the German AI stack — compact, local-first vision-language models, agentic systems, and LLM infrastructure. Research-grade, production-shipped, open-source-first.
(4× Critical)
on GitHub
for llama.cpp
German invoices tested
German-OCR-3 v0.2
Compact German document Vision-OCR — two editions (1.0 GB / 2.7 GB) on Qwen3.5, Apache 2.0. Zero hallucination on 200+ anonymized German invoices.
Research · Products · Infrastructure
🇩🇪 German AI Stack
German-OCR-3
live · 76⭐Local Vision-OCR for German documents. Apache 2.0.
German-Privacy-Shield
GDPRPII detection & redaction for German docs.
German-Layout
OSSDocument layout analysis for DACH forms.
German-TTS
WIPGerman text-to-speech — public release soon.
Deutschland-Assistent
comingAI assistant for German workflows.
🧠 LLM Infrastructure & Research
🛡️ Security AI
6 published CVEs · 10 advisories in triage
Responsible-disclosure security research via NVD, GitHub Security Lab, Patchstack and vendor programs. Targets include enterprise-grade systems at Siemens, Tutanota, Mailcow, Shopware, Pimcore, Einride and NousResearch.
| CVE | Severity | Type | Target | Year |
|---|---|---|---|---|
| CVE-2026-40492 | 9.8 Critical | Native RCE vector | SAIL image library | 2026 |
| CVE-2026-40493 | 9.8 Critical | Native RCE vector | SAIL image library | 2026 |
| CVE-2026-40494 | 9.8 Critical | Native RCE vector | SAIL image library | 2026 |
| CVE-2021-4455 | 9.8 Critical | Arbitrary File Upload · RCE | WordPress · Smart Product Review | 2021 |
| CVE-2021-24997 | 6.5 Medium | Auth / Info Disclosure | WordPress · WP Guppy | 2021 |
| CVE-2025-0990 | 4.3 Medium | CSRF | WordPress · I Am Gloria | 2025 |
- Siemens — KAS · linux-entra-sso (2 findings)
- NousResearch — hermes-agent (2 findings · incl. Critical)
- Tutanota — end-to-end encrypted mail
- Mailcow — dockerized mail server
- Shopware — commerce platform (2 findings)
- Pimcore — admin-ui-classic-bundle
- Einride — iam-go identity framework
- OSCP · Offensive Security Certified Professional
- CEH · Certified Ethical Hacker
- ISTQB · Certified Tester Foundation
- ISO 21434 · Automotive Cybersecurity (Vector)
Responsible disclosure only via NVD, GitHub Security Lab, Patchstack. No 0-day sales, no PoC leakage.
Academic work · Publications
AI for Engineering Systems
Technical University (TU) Darmstadt
Focus: Intelligent document processing, agent-based systems for civil engineering and industrial applications.
Cybersecurity (HDBW Munich)
Master's Thesis
"Manipulation of Neural Language Models — Security Risks and Defense Strategies for Trojaned AI Models"
- Verbesserung von Belohnungsmodellen durch Adversariale Trigger-Generierung Zenodo · 2024
- SecIDS-CNN: Advanced Convolutional Neural Network for Intrusion Detection Zenodo · HF · 2023
- SecIDS-CNN: Ein CNN für Intrusion Detection in Fahrzeugnetzwerken Zenodo · HF · 2023
- AI-Bilderkennungs-Klassifizierung GitHub · HDBW · 2021
What I use to build this
Advisory · Consulting · Research collaboration
Selectively taking on engagements where I can contribute deep technical work — agentic AI, document intelligence, LLM infrastructure, on-premise / DSGVO-compliant ML stacks, AI security.
Architecture reviews, model-selection, fine-tuning strategy, GPU sizing.
Joint work on VLMs, document AI, agentic systems, AI safety.