Skip to content

Job · Cohere

Software Engineer, GPU Infrastructure

EmployerCohere
LocationSeattle (remote)
FunctionTechnology & engineering
TypeFull-time
Pay (as disclosed)$175,000–$325,000 per year
Posted6 Nov 2026

About the role

Cohere seeks a Software Engineer, GPU Infrastructure to manage multi-cloud GPU and TPU clusters for its internal model training and serving workloads. The role involves deploying Kubernetes fleets, automating fault detection, and collaborating with cloud providers on hardware reliability. Candidates should have experience operating large-scale GPU clusters and writing software in Python or Go.

Summary written by GAR, Octum’s cognitive engine, from the employer’s posting. The full description is on the employer’s site.

More roles at Cohere

All Cohere roles →

Collected by Octum from Cohere’s careers site on 10 Nov 2026; Octum is not the employer and does not take applications. Spotted something wrong? Tell us.