Hanna van der Vlis
Hanna is an AI research engineer at Adyen - a payment processor processing over 40B transactions a year, where she focuses on deep learning research, training infrastructure, and building foundational models. She is passionate about solving large-scale engineering challenges and deploying AI to optimize complex systems and drive real-world impact.
Session
Pretraining foundation models on tabular and sequential data presents challenges that differ fundamentally from NLP or vision. This talk covers the key design decisions involved in building a payments foundation model trained on billions of transactions and trillions tokens: tokenisation of heterogeneous features, sequence construction, masking strategies and pretraining objectives for sequential tabular data, and the architectural trade-offs between hierarchical and language-model framings. Attendees will leave with transferable techniques for adapting self-supervised pretraining to non-text sequential data.