Research

Applied research on data governance, data products, and AI — published in peer-reviewed venues and turned into open source.

Google Scholar DBLP Entropy Data Research

Current focus

  • Automating data governance with generative AI
  • Bridging enterprise data products and LLMs through MCP
  • Open standards: Open Data Contract Standard (ODCS), Open Data Product Standard (ODPS)

Recent papers

Automating Data Governance with Generative AI

AAAI/ACM Conference on AI, Ethics, and Society (AIES) 2025 · with Linus W. Dietz, Arif Wider

Uses LLMs to evaluate natural-language governance rules against metadata. Generated 3.6× more warnings than human experts while catching all compliance concerns; 80% of AI-generated warnings were correct on review.

Paper
PDF
DOI
Data Product MCP: Chat with your Enterprise Data

arXiv preprint, 2026 · with Marco Tonnarelli, Filippo Scaramuzza, Linus W. Dietz

An MCP server that bridges enterprise data products and LLM-based assistants while preserving governance and access controls.

Paper
PDF
DOI

Most impactful papers during my time at University of Bamberg

Published 20+ peer-reviewed papers during PhD at Otto-Friedrich-Universität Bamberg (Dr. rer. nat., magna cum laude, 2017). Focus areas: BPMN/BPEL, workflow engine benchmarking, service composition.

GitOps: The Evolution of DevOps?

IEEE Software, 2022 · with Florian Beetz

Examines whether GitOps is truly an evolution of DevOps by analyzing and comparing their key elements, since neither has a standard definition. Identifies how GitOps extends DevOps practices through declarative, Git-centered continuous deployment.

Paper
DOI
BPMN 2.0: The state of support and implementation

Future Generation Computer Systems, 2018 · with Matthias Geiger, Jörg Lenhard, Guido Wirtz

Investigates how BPMN 2.0 implementers deal with the standard. Only three of 47 BPMS support the execution format defined in the spec, although all claim compliance. Tracks camunda BPM, jBPM and activiti over more than three years and finds feature support has plateaued at roughly half of BPMN's features.

Paper
DOI
A Pattern Language for Workflow Engine Conformance and Performance Benchmarking

EuroPLoP 2017 · with Jörg Lenhard, Oliver Kopp, Vincenzo Ferme, Cesare Pautasso

Captures common solutions for recurring problems in workflow engine conformance and performance benchmarking — test identification, benchmarking procedure validation, automatic engine interaction, and workflow execution observation — as a pattern language to help future benchmark authors.

Paper
PDF
DOI
Code Process Metrics in University Programming Education

SE Workshops (CEUR Vol-2308), 2019 · with Linus W. Dietz, Robin Lichtenthäler, Adam Tornhill

Applies behavioral code analysis (commit-based process metrics like churn and complexity trends) to student programming assignments to give actionable feedback beyond functional correctness, and to detect quality regressions early in coursework.

Paper
PDF
Teaching Clean Code

SE Workshops (CEUR Vol-2066), 2018 · with Linus W. Dietz, Johannes Manner, Jörg Lenhard

Presents a feedback-driven teaching concept for second- and third-year university students focused on clean and maintainable code, refined over more than six years of practice and recognized with the faculty's teaching award.

Paper
PDF