Work and Education

school Work Experience

Forschungsinstitut Bildung Digital (FoBiD)

  • Saarland University
  • July 2024 – July 2025
  • Saarbrücken, Germany

Scientific Employee

  • Developed a Java Spring Boot-based platform, schooltogo, to support teachers in creating and organizing their lessons efficiently.
  • Implemented a content-based recommender to help teachers organize their materials and lessons.
  • Used OpenAI’s text generation models to help teachers speed up their text writing process and to extract school competences from the official Federal state Curricula PDFs which are then imported into schooltogo.
  • Conducted interviews with educators to evaluate platform usability and inform future feature development.

Smart Enterprise Engineering (SEE) Group

  • Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI)
  • July 2023 – July 2025
  • Osnabrück, Germany

Researcher

Developed various recommender systems and improved DevOps practices for the two following research projects:

  • KUPPEL: an AI-based platform which used a sequential-based recommender to find the optimal learning path courses across two different German learning platforms.
  • YouCodeGirls: developed a platform which encourages young girls to learn programming, by tailoring the courses to their interests, e.g. sports, music.

Didactic Innovations GmbH

  • July 2021 – July 2024
  • Saarbrücken, Germany

Software Engineer

  • Developed the content creation platform, COLORS, to help companies internally train their employees. COLORS uses Java Spring Boot in the backend and stores all data in a PostgreSQL database.
  • Implemented a content-based recommender to guide the trainer through the existing courses on the platform.

school Research Experience

Prof. Dr. Vera Demberg's research group

  • Saarland University
  • August 2020 – February 2021
  • Saarbrücken, Germany

Scientific Employee

  • Developed a natural language generation model of bar chart descriptions in a few-shot setting.
  • Adapted an existing TensorFlow and Python codebase to work with a dataset of annotated bar chart descriptions.
  • Automatically measured the correctness and quality of the GPT-2 generated descriptions.
https://www.uni-saarland.de/lehrstuhl/demberg.html

Prof. Dr. Cas Cremers's research group

  • CISPA Helmholtz Center for Information Security
  • May – July 2019
  • Saarbrücken, Germany

Student Assistant

Proposed and analyzed multiple strategies for extending the Signal messaging protocol to multiple devices while preserving its security guarantees.

https://cispa.saarland/group/cremers/index.html

Secure and Privacy-Preserving Systems research group

  • Saarland University
  • May – August 2016
  • Saarbrücken, Germany

Student Assistant

  • Theoretical analysis of security protocols for smart homes and IoT.
  • Analysis of the WhatsApp encryption protocol.

book Teaching Experience

System Security research group

  • Saarland University
  • November 2017 – May 2018
  • Saarbrücken, Germany

Tutor

Organized and taught a weekly tutorial, which included the preparation and grading of minitests. Helped with exam preparation and grading.

https://cispa.saarland/group/rossow/news

Program of Psycho-Pedagogical Training, Level I

  • Department of Teacher Training, West University of TimiÈ™oara
  • October 2013 – August 2014
  • TimiÈ™oara, Romania

Student

  • Course work about Psycho-Pedagogical Training, how to organize lessons, and hands-on teaching experience.
  • Taught high school students introduction to programming.

CoderDojo

  • TimiÈ™oara Startup Hub
  • April – September 2013
  • TimiÈ™oara, Romania

Mentor

Helped young children learn how to program in Scratch

school Education

Saarland University

  • Prof. Vreeken
  • October 2014 – May 2019
  • Saarbrücken, Germany

Master in Computer Science

Thesis: How to be Grim: explaining data at different granularity levels using patterns and structure functions

  • Proposed two algorithms, Grim and Brim, which approximate Kolmogorov’s structure function. They find multiple, increasingly detailed explanations for a dataset, that provide a high-level view as well as an in-depth one.
  • Improved and extended an existing large codebase in C++.

West University of Timișoara

  • Prof. Istrate
  • October 2011 – July 2014
  • TimiÈ™oara, Romania

Bachelor in Computer Science

Thesis: The experimental analysis of several approximation algorithms

  • Implemented multiple approximation algorithms in Python for solving the Set Cover and Vertex Cover problems.
  • Compared the results with an optimum solution outputted by the mathematical optimization solver Gurobi.

West University of Timișoara

  • Prof. Acea
  • October 2011 – July 2014
  • TimiÈ™oara, Romania

Bachelor in Photography

Thesis: People and books

  • Course work about Art in general and Photography in particular.
  • The final thesis consisted of a series of photographs depicting people reading their chosen books.

psychology Technical Skills

Programming Languages

Advanced: Python, C++, C, LATEX, JavaScript, HTML, CSS, PHP

Intermediate: Bash, R, Java, C#

Packages/libraries

NumPy, Matplotlib, NLTK, Scikit-learn, Gensim, spaCy, Polyglot, Pandas

Machine Learning & NLP

Deep Learning (LSTM, Encoder-Decoder, Transformer) using TensorFlow (Keras)

Named Entity Recognition, Natural Language Generation

Other

Linux, Git, Bootstrap, Docker, MySQL

language Languages

🇷🇴 Romanian (Mother tongue)
🇬🇧 English (C1)
🇩🇪 German (B1)
🇫🇷 French (B1)
🇪🇸 Spanish (B1)

build Projects

NLG of Bar Chart Descriptions in a Few-Shot Setting

I created an approach for generating bar chart descriptions automatically. Since there are few datasets available for training which contain bar chart descriptions, a secondary goal was to do this with a small training set. Thus, I developed a natural language generation model of bar chart descriptions in a few-shot setting. This included adapting an existing TensorFlow and Python codebase to work with a dataset of annotated bar chart descriptions. I also proposed an automatic measure for computing the correctness and quality of the GPT-2 generated descriptions.

How to be Grim: explaining data at different granularity levels using patterns and structure functions

The focus of my Master's Thesis was finding out whether it is more beneficial to analyse a dataset via multiple models, instead of just a single one. By using Kolmogorov’s structure function we can not only order these on how well they explain the data, but also identify those that are worth looking into. Thus, I created two algorithms, Grim and Brim, for finding these models and through many experiments showed that the algorithms output high-quality, high-level, as well as in-depth explanations.