Diana Kozachek

Diana Kozachek

Diana Kozachek

Wissenschaftliche Mitarbeiterin

Schwerpunkte

Agentic AI und Multi-Agenten-Systeme

Personalisierte Lernsysteme und AI-gestütztes Scaffolding

Large Language Models und RAG-Architekturen

Foresight, Szenarienentwicklung und Zukunftsstudien

NLP und datengetriebene Analyse großer Textkorpora Human–AI Interaction

Empirische Nutzerstudien

Experimentelle Forschung

Ausbildung
  • University of Saint Gallen, Switzerland (HSG), PhD Student in Information Systems
    Fokus on: Developing Agentic AI Systems to support Life-long Learning

  • Freie University of Berlin, Germany, Master’s degree in Future Studies and Computer Science
    Thesis on: The Visionary Machine: Can GPT-3, GPT-3,5, Or GPT-4 Generate Human-Like Future Scenarios?

  • University of the Arts Berlin, Germany, Bachelor’s degree in Business Communication Science 
    Thesis on: Design Thinking in Social Startups: Analysis of Kiron NGO
Berufserfahrung
  • Doctoral Researcher, Institute for Information Systems, University of St. Gallen, Switzerland February 2025 – Present
    - Development and analysis of a Multi-Agent-System to personalize learning of problem-solving skills in students.
    - Development of a Multi-Agent-Recommender-System for Executive Education program admissions.

  • Applied Scientist in R&D, Volkswagen Group China, V-Labs, Beijing, China July 2024 – February 2025
    Developed light-weight LLM solutions for the Strategy Steering Division.
    - Implementation of LLMs in legal decision-making processes by building a multi-lingual RAG system.
    - Setup of an automated, multi-lingual content scanning and scraping system on WeChat to support the semiconductor strategy 2026 across the entire supply chain.
    - Subsequent creation of scenarios for the semiconductor strategy 2026.

  • Applied Scientist in R&D, Fraunhofer Institute for Innovations and Systems Research (ISI), Berlin, Germany February 2021 – September 2023
    - NLP-based survey analysis of a nation-wide study on sleep patterns.
    - Evaluation of bio-inspired optimization algorithms for sleep-quality enhancing light systems.
    - Setup of automated web-content scanning and reporting for consumer electronics (Bosch and Siemens).

  • Guest Researcher & Lecturer, Metaverse XR Lab, National Taipei University of Technology, Taipei, Taiwan July 2022 – October 2022
    - Conceptualized and executed a user study on remote collaboration in VR.
    - Analyzed biophysical data (Electro Dermal Activity and Eye Tracking) to study stress factors in virtual work environments.
    - Planned, applied for funding and published two peer-reviewed studies.
    - Led a team of five undergraduate and graduate students.
    - Taught a beginner data analytics course with Python for 12 graduate students.

  • Graduate Research Assistant, Fraunhofer Institute for Production Systems and Design Technology (IPK), Berlin, Germany April 2020 – March 2021
    - Quantitative research on virtual product creation and management, focusing on light signals for autonomous public transportation vehicles.
    - Conducted user studies with eye-tracking and data analysis using machine learning algorithms to enhance visibility of visual signals and derive design rationales.

  • UX/UI Concepter, Scholz & Friends, Berlin, Germany August 2017 – July 2019
    - Developed and tested UX/UI applications for Audi, PayPal, Hyundai, Daimler, and public sector clients including the European Commission.
north