researcher / ai engineer / consultant

Arman
Arzani

I build AI systems that find insights in messy data, as a researcher at the University of Duisburg-Essen, and as a consultant developing AI concepts for major facility management companies in Germany.

loc
essen, de
edu
phd candidate, uni due
field
data mining · ml · nlp
Portrait of Arman Arzani

I'm a Persian-German computer scientist and currently a PhD candidate at the University of Duisburg-Essen, where I specialize in knowledge discovery, data mining in large datasets, and LLMs. I got into NLP in 2020, started tinkering with transformers, and had too much fun to ever stop. The ML systems I was building back then have since grown into full-scale solutions with frontier models. Since the start of my career, I have been building AI systems for research and industry, including retrieval pipelines, business analytics, AI agents, and training LLMs for real-world applications.

Today I consult for some of Germany's largest facility management companies, developing AI concepts and strategy, building end-to-end AI solutions, and running AI-enablement workshops for the people who work with them every day.

currently
PhD candidate, University of Duisburg-Essen
consulting
AI concepts & strategy for facility management
ai engineering
LLM fine-tuning, RAG, agentic systems, evals
experience
6+ years building software
based in
Essen, Germany
pythonjavatypescriptjavascriptangularnext.jsnode.jspostgresqldockerelasticsearchjenkinswordpress
2025

A Hybrid Approach for Mining the Organizational Structure from University Websites

Arman Arzani, Theodor Vogl, Marcus Handte, Pedro Marrón

Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR, INSTICC SciTePress

#data-mining#knowledge-discovery#organizational-structure
2024

Discovering Potential Founders within Academic Institutions

Arman Arzani, Marcus Handte, Pedro José Marrón

International Journal of Data Science and Analytics · ISSN 2364-4168

#data-science#analytics#innovation
2023

Exploiting Topic Modelling for the Identification of Untapped Scientific Collaborations

Arman Arzani, Marcus Handte, Matteo Zella, Pedro José Marrón

2023 the 7th International Conference on Information System and Data Mining (ICISDM), pp. 73–81, Association for Computing Machinery, Atlanta, USA · ISSN 9798400700637

#topic-modelling#scientific-collaboration#data-mining
2023

Challenges in Implementing a University-Based Innovation Search Engine

Arman Arzani, Marcus Handte, Pedro José Marrón

Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KDIR, pp. 477-486, INSTICC SciTePress · ISSN 2184-3228

#innovation#search-engine#knowledge-discovery
2023

Discovering Potential Founders Based on Academic Background

Arman Arzani, Marcus Handte, Matteo Zella, Pedro José Marrón

Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KMIS, pp. 117-125, INSTICC SciTePress · ISSN 2184-3228

★ best poster award#academic-background#innovation#knowledge-management
Homemade bagel toast
Travel photo
Clay sculpture
My two cats

Away from the keyboard, sculpting with polymore clay is my offline activity. Working with clay is the opposite of working with large language models, slow, physical, and hallucinatory in its own way, which is exactly why I keep doing it. It helps me stay grounded, and reminds me that some things are best done by hand.

The rest of my time goes to traveling, films that reward attention, hanging out at tennis courts, and two cats who tolerate all of the above.

/ on repeat

current favorites

  • 28 Days Later Main ThemeYoung Fathers
  • Don't SingData, Benny Sings
  • I Appear MissingQueens of the Stone Age

/ weather @ essen

fetching data…

For research collaborations, consulting inquiries, or anything else, email is the fastest way to reach me.

copied to clipboard