Alma Liezenga
Utrecht, the Netherlands
About me
Welcome to my personal website! My name is Alma Liezenga and I am a data enthusiast from the Netherlands. This website gives a brief overview of my professional background, projects I have worked on, and blog posts I have written.
I have previously studied Information Sciences (B.Sc.) at Utrecht University (2016-2019). After a gap year to take on a board position at my student association, I followed the master's program in Engineering & Policy Analysis at TU Delft (2020-2022). I supplemented this degree with a specialization in Machine Learning at Lund University (Sweden) and a graduation project with Spare Labs Inc. on the impact of on-demand transit on access equity in the Minneapolis-St.Paul metropolitan area (Minnesota, USA).
After completing my education I took a job at Trunkrs as a data scientist. My responsibilities there included forecasting, GIS, and machine learning projects. Simultaneously, I started volunteering with Fruitpunch AI. Their AI for good projects offered a meaningful addition to my day-to-day work. Currently, I am working at TNO as a computer vision scientist. At TNO, we conduct research to foster innovation. I particularly enjoy working on the topics of generative AI, synthetic data for deep learning, and AI for security.
Writing
I love to share some of my ideas and data-related work in writing. Here's an overview of my blog posts:
- Evaluating LLaMA 3 for political sycophancy, part of my capstone project for BlueDot Impact's AI safety fundamentals course.
- Two new datasets for evaluating political sycophancy in LLMs, part of my capstone project for BlueDot Impact's AI safety fundamentals course (also on LessWrong).
- Hoe extraparlementair is Schoof I? De politieke achtergrond van 11 kabinetten (NL)
- Combining IMDb data with the Bechdel Test my submission for the Y42 hackathon.
- Het geschenkenregister gevisualiseerd (NL)
- Leveraging Large Language Models to make businesses around the world more sustainable! Co-written for Fruitpunch AI
- Cabinet Members of the Netherlands: a dataset
- Top 2000: web scraping, interactive plots & holiday fun
- How to Hire Women: a Guide to Inclusive Vacancies
Occasionally, I manage to write down my thoughts and findings in a more coherent way. This has resulted in the following scientific publications thus far:
- Liezenga, A. M., Verma, T., Mayaud, J. R., Aydin, N. Y., & van Wee, B. (2024). The first mile towards access equity: Is on-demand microtransit a valuable addition to the transportation mix in suburban communities?. Transportation research interdisciplinary perspectives, 24, 101071.
- Ruis, F. A., Liezenga, A. M., Heslinga, F. G., Ballan, L., Eker, T. A., den Hollander, R. J., van Leeuwen M. C., Dijk, J., & Huizinga, W. (2024). Improving Object Detector Training on Synthetic Data by Starting With a Strong Baseline Methodology. arXiv preprint arXiv:2405.19822. Presented at SPIE Defense + Commercial Sensing.
Public speaking
I enjoy sharing what I've learned with an audience. This is a non-exhaustive list of places I've spoken. I am always interested in opportunities to talk about my professional and personal work so do not hesitate to reach out if you're organizing a conference or meet-up.
- Presenter at the VI International Seminar on Contemporary Mobilities by the Latin American Chapter of the International Network for Transport and Accessibility in Low-Income Communities on "The first mile towards access equity: Is on-demand microtransit a valuable addition to the transportation mix in suburban communities?"(19-09-2024).
- Moderator at PyData Eindhoven 2024 (11-07-2024).
- Guest on the Inspiring Computing podcast S01E11, talking about The New Urban Equation: Data, Transit, and Sustainable Cities" (27-05-2024).
- Presenter at SPIE Defense + Commercial Sensing conference on "Improving Object Detector Training on Synthetic Data by Starting With a Strong Baseline Methodology", in Washington D.C. (25-04-2024).
- Host at exec(ut) conference 2024, for approximately 300 IT students, in Utrecht (27-03-2024).
Projects
Besides my professional work I enjoy organizing events and contributing to AI for good projects. This is a (non-exhaustive) list of projects I've contributed to:
- PyData Eindhoven's 2024 edition as an organizer
- Fruitpunch AI's AI for impact assessment challenge as an AI engineer for good
- Fruitpunch AI's AI for preterm babies challenge as an AI engineer for good
- Onehot – data science meetups as an organizer
- exec(ut) 2019 as committee chair
- exec(ut) 2018 as commissioner of external affairs