Great to have you here. I am Siiri, and here you can find a bit more about me and what I have worked with.
I worked as a research assistant in TurkuNLP from May 2024 to December 2025. During my time there, I had the pleasure to contribute to two different projects.
The most recent one was work on Finnish embeddings, where I...
- preprocessed raw-ish data into a new dataset
- embedded and indexed the data
- run various tests on openly available embedding models
- visualized results
- annotated samples of the data
to name a few things. Not all work is openly accessible due to the restricted nature of the data.
The first project I got to contribute to was published as OCR Error Post-Correction with LLMs in Historical Documents: No Free Lunches (Kanerva et al., 2025). In the related repository, some of my work can be accessed. Again, due to data distribution restrictions, not all my contributions are publicly available. My tasks included but were not limited to
- data preprocessing
- exploratory data analysis
- refactoring code
- annotation.
I am currently aiming to finish my master's degree in computer science by the end of Spring 2026. I am also planning to start to work on an annotation tool, a full-stack project that complements my university studies. (The link will be available here shortly.)
Oh, and if you found your way here through the link in my CV: I wrote that document in HTML and CSS. Here I have just replaced my information with some mock data. (Who knows, maybe one day this CV turns into an interactive webpage?)