On-Prem AI Platform Engineer — Flexible Hours

SatoshiLabs Group; Jackson (MS); Vor Ort; USD 110.000 - 170.000; Vollzeit

Benefits dieser Stelle

Option to be paid in bitcoin

Flexible working hours

Budget for professional development

Open, friendly culture

Renovated offices with gym and games

On-site parking and MultiSport card

Zusammenfassung

Trezor is building its own AI inference infrastructure in-house, hosting a multi-GPU server in our data center. We seek an AI Platform Engineer to own the hardware, OS, model serving stack, and user enablement, delivering end-to-end control for benchmarking, deployment, and optimization.

You will collaborate with IT and developers to wire agents, maintain API access, and write internal guides that enable self-service, while advancing the open-weight model ecosystem and ensuring secure, fast

Qualifikationen

• Coding experience required to build and maintain AI infra.
• Infrastructure as Code skills (Ansible, Terraform) to automate work.
• Linux system administration and kernel-level troubleshooting.
• Networking, storage, containers, systemd, and IT collaboration.
• English fluency for daily work; Czech is a plus.

Aufgaben

• Own GPU server lifecycle: OS, storage, networking, and observability.
• Develop and run the inference stack with vLLM/SGLang/LiteLLM, Open WebUI front end.
• Benchmark models, evaluate releases, and optimize for quality, throughput, latency.
• Manage quotas, routing, and cost reporting across teams; assist cloud usage as needed.
• Support engineers, create onboarding guides, run internal workshops.

Kenntnisse

Coding experience

IaC skills

Linux admin

Networking

Troubleshooting

Docker/Kubernetes

CI pipelines

Tools

Docker

Kubernetes

CUDA

nvidia-smi

Jobbeschreibung

Trezor is building its own AI inference infrastructure in-house, hosting a multi-GPU server in our data center. We seek an AI Platform Engineer to own the hardware, OS, model serving stack, and user enablement, delivering end-to-end control for benchmarking, deployment, and optimization.

You will collaborate with IT and developers to wire agents, maintain API access, and write internal guides that enable self-service, while advancing the open-weight model ecosystem and ensuring secure, fast