Distributed Systems Engineer

What you'll do

• Develop the routing layer that selects machines for inference requests and handles demand exceeding available capacity.
• Build and operate systems that scale a multi-cloud GPU fleet as demand changes.
• Work on Relace’s Rust services and Kubernetes services managing the fleet.

What they ask for

• Around five or more years building production systems where latency and failure behavior mattered.
• Strong knowledge of a systems language such as Rust, C++ or Go.
• Ownership of a production system that ran under load, and detailed experience debugging a production failure.
• Kubernetes experience extending it, particularly on low-reliability infrastructure.

Nice to have

• Project experience with a scheduler or load balancer; work with spot or preemptible capacity.
• Exposure to AI serving systems.

Tools

Rust, Kubernetes, C++, Go

AI focus

Build infrastructure for globally distributed AI inference, including request routing, capacity management, and fleet scaling.