Teaching
Having the opportunity to teach was, and still is a privilege. Indeed, it was an important reason for deciding to pursue a PhD degree. Throughout the years, I have taught many different courses. A selection of these are discussed below.
Natural Language — NLP
I have taught the NLP course at DSV since 2022. The course was completely new, and I was part of the team designing it. I helped design the curriculum, several lectures, and also the lab assignments. We are constantly finding ways to improve the course, and I'm proud of the consistently positive feedback we find in our course evaluations.
Internet Search Techniques and Business Intelligence — ISBI
I have taught parts of the ISBI course since 2022. This course is aimed at bachelor's students at DSV, and focuses on techniques for searching and analyzing information on the internet. The course starts with traditional information retrieval, then introduces core NLP concepts, and ends with an introduction to modern approaches to internet search using large language models.
Human-AI Interaction — MAID
I have been involved in the MAID course since 2025. My contribution is an introductory lecture about machine learning and how modern deep learning approaches work from a technical perspective. The course is taken by bachelor students focused on interaction design, and the lecture provides a technical but accessible overview of how AI actually works.
Selected Course Materials
Here is a selection of course material I have created over the years.
| Course | Type | Title | Materials |
|---|---|---|---|
| NLP | Lecture | Classification in NLP | slides and recording |
| NLP | Lecture | Deep Learning | slides and recording |
| ISBI | Lecture | Opinion Mining and Sentiment Analysis | slides and recording |
| ISBI | Lecture | NLP for information retrieval | slides and recording |
| MAID | Lecture | Machine Learning and Deep Learning | slides and recording |
Supervision
As of 2026, I've supervised 16 master theses and three bachelor theses. These theses have explored a wide range of ML and NLP topics, from gender bias detecting in LLMs, through privacy-preserving methods for NLP, to NLP-based analysis of Polish rap lyrics.
| Year | Title | Student(s) |
|---|---|---|
| 2026 | Bars from the Blocks: Computational Discourse Analysis of Polish Rap (1994–2026) | Zuzanna Anna Chodorowska |
| 2026 | Anonymization of hierarchical telecommunication data | Soeren Christopher Raymond |
| 2026 | Requirement-level compliance assessment in QA workflows: A retrieval-augmented and evidence-audited LLM prototype | Tatiana Mikhailova |
| 2026 | Multimodal Emotion Monitoring under Data Sparsity: Feasibility of Integrating Continuous Wearable Signals with Intermittent Speech | Yixuan Li |
| 2026 | Explainable and Data-driven AI-Methods for Predictive Maintenance | Monir Hasani & Helin Sayed Youssef |
| 2026 | Prompt Language as a Bias Modulator | Tianhua Chen |
| 2025 | From Prompt to ICD-10: Evaluating Prompt-Based Decoder Large Language Models for ICD-10 Coding Using a Swedish Clinical Dataset | Diana Amin & Katarina Karlin |
| 2025 | Evaluating LLMs for Gender Bias in Chinese Language: A Comparison of Multilingual and Chinese-based Models | Donghui Lyu & Liangliang Ma |
| 2025 | Evaluating LLMs for Bias in Non-English Languages - A Persian Perspective | Mohammad Soroush Bagheri |
| 2025 | NLP Slang Time Machine: Enhancing Model Accuracy for Slang Integration | Emilio Estevez Gustavsson & Adrian Höglund |
| 2024 | Subword segmentation in highly compounding and morphologically complex languages | Maximilian Törnqvist |
| 2024 | Assessing a Swedish Automatic Speech Recognition Model for Finland Swedish | Amanda Gisslén & Jessica Machutta Widengren |
| 2024 | Enhancing the debugging procedure of Petrel Software (SLB) using call stacks | Elza Cherian |
| 2024 | In-Vehicle Privacy Using AI | Iyad Al Khatib |
| 2023 | Towards Building Privacy-Preserving Language Models: Challenges and Insights in Adapting PrivGAN for Generation of Synthetic Clinical Text | Atena Nazem |
| 2023 | Mapping adverse drug reactions to patient-reported symptoms using NLP | Heaven Gebremeskel Weldeslasie |
| 2022 | Evaluation of Post Hoc XAI Models in Explaining Automatic ICD-10 Classifications of Swedish Discharge Summaries | Hjalmar Davidsen & Alexander Dolk |
| 2022 | Bias in De-identification Systems | Xiaoman Tang |
Are you a student with a thesis idea? Contact me!