Scholar
Lennart Purucker
Google Scholar ID: x_HyTt0AAAAJ
PhD Student
Citations & Impact
Citations
469
H-index
8
i10-index
7
Publications
20
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- In 2024, mainly worked on TabPFN, a foundation model for (small) tabular data. Participated in Kaggle's AutoML Grand Prix as the leader of the 'AutoML Grandmasters' team, where they scored a very close (1 point) second place with AutoGluon and TabPFN, winning $20,000. In early 2025, published the newest version of TabPFN in Nature. For the rest of the year, embarked on a new mission: creating reliable, rigorous benchmarks for tabular data. Started with a living benchmark for IID tabular data – TabArena.
Research Experience
- Worked as a research assistant at the University of Siegen from November 2021 to August 2023, focusing on ensemble learning for Automated Machine Learning (AutoML) and recommender systems. Interned as an applied scientist at AWS as part of the AutoGluon team from August 2023 to November 2023.
Education
- Bachelor's degree in Applied Computer Science from DHBW Stuttgart in 2019; Master's degree in Computer Science from RWTH Aachen in 2021.
Background
- Research interests: Artificial Intelligence, Automated Machine Learning, Ensemble Learning, Deep Learning, Meta-Learning (for small data). Main focus on tabular data (e.g., Excel sheets), but also works on vision, text, and time series data.
Miscellany
- Community involvement: Reproducibility Chair at the AutoML Conference 2023, 2024, and 2025; Co-organizer of the AutoML Seminar; Co-organizer of workshops related to tabular data and foundation models (ICML'25, EurIPS'25); Developer of AutoGluon (Tabular); Member of the OpenML Team (Python API); Core Maintainer of TabArena; Invite-only Community Seminars: Dagstuhl (24024, 24082, 25182), Shonan (223).
