Incoming PhD candidate at VU Amsterdam developing multimodal ML models to assess wellbeing from video, speech, and behavioural signals. Econometrics & Data Science graduate with a focus on NLP, affective computing, and reproducible pipelines.
I'm a quantitative graduate from the University of Amsterdam, where I completed my MSc in Econometrics with a strong focus on machine learning and statistical modeling. My thesis on NLP-driven market sentiment analysis was awarded a 9/10 and won the REmagine thesis prize in Responsible Digital Transformation.
I'm experienced in building end-to-end Python pipelines for data collection, preprocessing, and model inference, with a strong emphasis on robust evaluation and reproducibility. Starting April 2026, I will be pursuing a PhD at Vrije Universiteit Amsterdam on Sentiment Analyses for Wellbeing, developing multimodal ML models to dynamically assess human wellbeing from video data, integrating facial expressions, speech, and behaviour.
Extensive quantitative training in econometrics, machine learning, and data science at the University of Amsterdam.
End-to-end machine learning projects with emphasis on reproducibility, evaluation, and real-world impact.
Multimodal pipeline for audio-visual emotion classification, demonstrating how evaluation methodology (random splits vs. actor-wise vs. external validation on CREMA-D) affects conclusions about model performance.
Award-winning thesis building a reproducible pipeline for news collection, preprocessing, and transformer-based sentiment inference using Hugging Face.
Combining research, teaching, and industry experience across ML, finance, and education.
Developing multimodal machine learning models to dynamically assess human wellbeing from video data, analysing facial expressions, tone, behaviour, and speech. Addressing methodological challenges including data quality, bias, and model validity, integrating data from the Netherlands Twin Register alongside open-source datasets.
Taught weekly tutorials to first-year BSc Sociology students in Introduction to Statistics. Guided SPSS-based analyses and helped students interpret results, adapting materials to their needs.
Client-facing role translating complex investment information into clear guidance, applying analytical reasoning to help customers make informed decisions.
Diagnosed learning gaps and tailored 1:1 tutoring sessions, translating complex concepts into clear explanations with targeted exam practice.
A strong quantitative foundation paired with modern ML frameworks and engineering tools.
Interested in collaboration, research opportunities, or just want to chat about ML? Reach out.