описание
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.
задачи
- 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.
- Help guide compute spending decisions. You understand cost per FLOP across providers and hardware, and you hate wasted compute
требования
- 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
- Будет плюсом: 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
условия
- 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.
- Compensation Range: €100K - €210K