Research Scientist, Foundational Data Science

Who We Are

Foundation models transformed text and images. Structured data – the largest and most consequential data format in the world – stayed untouched. Tables run every clinical trial, every financial model, every scientific experiment, every business decision, and no one had built a foundation model that truly understood them.

We pioneered tabular foundation models and are now the world-leading organization in structured-data ML. Our TabPFN v2 model was published as a Nature cover story and set a new state‑of‑the‑art for tabular machine learning. Since its release we have scaled model capabilities 20×+, passed 3.5 M+ downloads and 7 500+ Git Hub stars, and are seeing accelerating adoption across research and industry – from detecting lung disease with Oxford Cancer Analytics to preventing train failures with Hitachi to improving clinical‑trial decisions with Boston Gene.

We’re scaling tabular foundation models to millions of rows, thousands of features, real‑time inference, and entirely new data modalities, while building the infrastructure to run them in production across some of the most demanding industries on earth. These are open problems no one else is working on at this level.

We’re a small, highly selective team of 30+ engineers, researchers, and GTM specialists, with backgrounds spanning Google, Apple, Amazon, Deep Mind, Meta, Microsoft Research, G‑Research, Jane Street, Goldman Sachs, and CERN. We are led by Frank Hutter, Noah Hollmann, and Sauraj Gambhir, and advised by world‑leading AI researchers including Bernhard Schölkopf and Turing Award winner Yann LeCun. We ship fast, do top‑tier research, and hold each other to an extremely high bar.

In 2025 we raised €9 m pre‑seed led by Balderton Capital, backed by leaders from Hugging Face, Deep Mind, and Black Forest Labs. The next phase of growth is here, making this an ideal time to join.

• Invent and build frontier tools that extend TabPFN, including its thinking, scaling, and agentic capabilities, and the new methods that let one model generalise across the full landscape of data‑science problems. This is the most open‑ended part of the work and grows over time.
• Set the research direction by deciding which model capabilities and benchmarks are worth pursuing, choosing what is worth solving rather than optimising a score someone else set.
• Bring in external research and real customer needs to shape new model and tooling directions, and publish frontier results that move the field forward.
• Build trustworthy benchmarks from the structured data behind real, high‑impact problems, so the team optimises for real‑world performance rather than one leader board.
• Faithfully implement the baselines and competitor models that set the gold standard of applied data science, giving the team a read on where TabPFN leads and where there is room to improve.
• Build an automated, agentic pipeline with a human in the loop so this data and benchmark foundation scales to far larger volumes without losing rigor, itself a genuinely new tool.

• You have solved data‑science problems across many domains and datasets to a high standard, optimising for strong performance across a whole suite of tasks rather than the single best score on one.
• You work undogmatically across the ML toolbox, including getting strong results with gradient‑boosted trees (such as XGBoost) and not only with deep learning.
• You understand the common categories of dataset defects (leakage, label noise, distribution shift, duplication, mislabelled targets, and similar) and why each corrupts a training or benchmark signal.
• You are energized by foundational work, valuing the dataset and benchmark bedrock as much as the frontier tooling, and you have taken on hard problems others passed over.
• You thrive as a senior individual contributor in an ambiguous, early‑stage, low‑process environment. You are opinionated on best practice in Data Science and can make good judgement calls on approaches to complex problems.

• Experience building or extending evaluation harnesses, benchmark suites, or experiment frameworks…