Hello! I am Anurag Garg, a Research Scientist at Prior Labs, where I work on advancing deep learning for structured data, including tabular and time-series data. I graduated from the University of Freiburg with a Master's in Computer Science, specializing in Artificial Intelligence, under the supervision of Prof. Frank Hutter, with an overall grade of 1.2, which corresponds to excellent in the German grading scale.
My master's thesis introduced Real-TabPFN, an extension of TabPFN — a transformer-based foundation model for tabular data — that achieves state-of-the-art results on tabular benchmarks. It was published at FMSD @ ICML 2025, and is now the default model in the TabPFN API.
During my tenure at Deloitte, I focused on data analysis and model building.
In my free time I enjoy exploring new developments in AI/ML, surfing Academic Twitter, cooking and going on hikes in the beautiful Black Forest.
