- Based in Stockholm
- Background in statistics and machine learning
- Currently working with logistics network optimization at H&M Group
Connect with me via LinkedIn in my profile!
- MSc in Statistics and Machine Learning, Linköping University 2025
- BSc in Statistics and Data Analysis, Linköping University 2023
Explored the use of deep learning, specifically Transformer-based models, for classifying femoral fractures using fusion of electronic health records and radiographic images. Built and compared 5 different models with 2 models using attention-based fusion; unimodal tabular, unimodal vision, multimodal late fusion, multimodal intra-attention fusion, multimodal inter-attention fusion. Implemented MC Dropout to account for prediction uncertainties per image and aggregated predictions to patient-level using uncertainties. Download and read the thesis here.
Analyzed the impact of digital marketing spend across multiple channels through mix marketing modeling on new customer acquisition for a fintech company. Built and compared XGBoost models with various transformations to capture carryover effects and seasonal patterns. Used SHAP values to interpret non-linear relationships and provided recommendations for optimal marketing budget allocation across channels. Download and read the thesis here (in Swedish).