Infra Engineer (Zurich)

Aioniclabs — Switzerland · Posted ~22 hours ago

🔓 Log in to save this job, tailor your resume & track your apply process — 7 days free, no card needed.

Log in to add to target list

Description

Full-time · On-site in Zurich, Switzerland · Founding team Aionic Labs is building the operational intelligence layer for the physical economy. With OpenTSLM (ICML '26) we introduced Time-Series Language Models that perceive signals, connect them with text, and explain their decisions. Building the foundation for model training, fine-tuning and inference on our large-scale GPU clusters. Shipping for the industry that demands infrastructure that does not break. You will own it. What you’ll do Build and operate our training platform: distributed multi-node GPU training, job orchestration, experiment tracking, model registry, and artifact management across hundreds of thousands of H200-hoursDesign our model fine-tuning and inference platform, deployable on any cloud and on-premOwn the compute and data backbone behind TimeNet: pipelines that move, version, and serve datasets as our corpus grows from millions to trillions of datapointsMake the platform reliable and reproducible for regulated deployments in healthcare, industrial, and energy settings, with the observability, security, and cost controls to matchWork directly with the founders and our research consortium of 50+ researchers across Stanford, ETH Zurich, Meta, AWS, and Google What we look for Strong engineering skills in Python and modern infrastructure tooling (containers, K8s, infrastructure-as-code, workload management, orchestration). You ship systems that run unattendedExperience running large-scale GPU training, fine-tuning or high-throughput model serving in productionCloud fluency, CI/CD, workflow orchestration, and production-grade observabilityA bias for reliability and cost-efficiency: you measure, profile, and remove bottlenecksDrive to build a category-defining company from Europe, in person, in Zurich Nice to have Distributed training frameworks (PyTorch) and inference optimization (vLLM, TensorRT)ML platform components: model registries, data lineage, experiment trackingOn-prem or regulated-industry deployments What we offer A founding-team seat on a frontier problem, backed by the SPRIND Next Frontier AI Iniative (up to €26.5M)Real compute: hundreds of thousands of H200-hours to run ideas at scaleCompetitive salary and meaningful equityAn office in Zurich and a team that ships, backed by the SPRIND Next Frontier AI Initiative with up to €26.5 million in funding. Website: https://www.aioniclabs.ai Application Form: https://aioniclabs.notion.site/team-careers