Infrastructure Engineer

Foundation models transformed text and images. Structured data - the largest and most consequential data format in the world - stayed untouched, until now. What LLMs did for language, we're doing for tables.

We pioneered tabular foundation models: TabPFN v2 was a Nature cover story, has passed 3.5M+ downloads and 7,500+ GitHub stars, and runs in production from detecting lung disease with Oxford Cancer Analytics to preventing train failures with Hitachi. The hardest problems - millions of rows, real-time inference, entirely new modalities - are still open, and no one else is working on them at this level.

We're a small, highly selective team of 40+ with backgrounds from Google, DeepMind, Meta, Apple, Amazon, Jane Street, and CERN, led by Frank Hutter, Noah Hollmann, and Sauraj Gambhir, and advised by Bernhard Schölkopf and Turing Award winner Yann LeCun.

In July 2026, less than 18 months after our €9M pre-seed, we joined SAP as an independent frontier AI lab - same team, mission, and open-weights models, now backed by more than €1 billion over four years.

We spend tens of millions per year on GPU compute to train tabular foundation models. That's not a target, it's what we're running today, and it's growing. The person who owns this infrastructure makes decisions worth millions of dollars: cluster architecture, scheduling efficiency, provider strategy, hardware selection. A wrong call costs six figures.

Today we run on Slurm across multiple clusters. We're scaling to multi-cluster, multi-provider infrastructure using SkyPilot and evaluating new hardware generations as they come online. You own the full stack, from cluster orchestration and cost optimization to CI/CD, automation, and the tooling that keeps researchers moving fast. You work directly with the research team and understand what they're doing well enough to make infrastructure decisions that actually help them.

What you'll work on:

• Own and evolve multi-cluster GPU infrastructure, from Slurm and SkyPilot on GCP today to multi-provider orchestration and new GPU generations tomorrow. Drive architecture, automated provisioning, workload scheduling, reliability, cost optimization, and capacity planning against growing compute demands

• Build the developer productivity layer: CI/CD pipelines, automated testing, reproducible environments, experiment tracking, and internal tooling that keeps research iteration speed high

• Improve infrastructure reliability through monitoring, observability, infrastructure as code, and automation. Reduce operational toil and make complex systems easier to manage.

• Own the compute budget. You understand cost per FLOP across providers and hardware, and you hate wasted compute

Tech stack: Slurm, SkyPilot, GCP (+ AWS and Azure to come), Docker, Terraform, GitHub Actions, Python, uv, wandb

You may be a good fit if you have:

• 3+ years building and operating production infrastructure at scale, ideally involving compute clusters, distributed training systems, or HPC

• Hands-on experience with cluster management and workload orchestration systems such as Slurm, Kubernetes, or similar technologies. You've debugged scheduling failures and operated infrastructure where downtime has real cost

• Experience designing and maintaining CI/CD pipelines, automated testing, and reliable build and deployment workflows

• Strong systems engineering fundamentals: Linux, networking, storage, and distributed systems. You're comfortable debugging complex infrastructure problems across the stack.

• Strong Python skills and enough familiarity with training to profile a run and tell whether the bottleneck is data loading, communication, or compute

• Track record of making infrastructure decisions that measurably improved reliability, developer productivity, or cost efficiency

• Strong AI tooling skills. You use Claude Code, Cursor, or similar fluently to move fast without sacrificing quality

Bonus:

• Familiarity with GPU infrastructure, PyTorch, GPU profiling, and distributed training

• Experience operating at tens-of-millions-scale compute spend

• Multi-cloud or hybrid HPC/cloud infrastructure experience

• Experience scaling from single cluster to multi-cluster orchestration

• Background building experiment tracking, model registries, or ML pipeline tooling

You'll work alongside researchers and builders who hold themselves to a very high bar - in the quality of their work and in how they work with each other. We move fast and still take the time to do things right.

Our teams are based in Berlin, Freiburg, and New York - when you're working on something as hard as TabPFN, being in the same room matters. But great people come from everywhere, and in exceptional cases we're open to remote, which usually means frequent travel to one of our offices. Wherever you're based, the whole company comes together regularly for offsites to build and celebrate together.

The best products and teams are built by people with a wide range of perspectives and backgrounds. We welcome applications from all identities and walks of life - especially if you've ever felt discouraged by "not checking every box" - and provide equal opportunities regardless of gender, sexual orientation, origin, disability, or any other trait that makes you who you are.

We care about how your data is handled - see our Recruiting Data Privacy page