The opportunity
Cohere is the leading security-first enterprise AI company. We build cutting-edge foundation AI models and end-to-end products that are designed to solve real-world business problems.
What you'll do
Own the platform that schedules and runs training, evaluation, and: reinforcement-learning jobs on Kubernetes: who gets capacity, what happens when a job fails, and how a researcher finds out whether the platform caused it.
Build the networking and runtime isolation those workloads need, including agent and sandbox environments.
Work with the data infrastructure team on storage for checkpoints and other training artifacts.
Troubleshoot production failures and performance problems, and leave the: system easier to operate than you found it.
Build automation, infrastructure-as-code, and observability so the platform stays maintainable.
Run a job platform, batch system, or Kubernetes controllers in production.: Scheduling, queues, and multi-tenant fairness are central to this work.
What they're looking for
- Shown distributed systems judgment: retries, partial failure, and knowing what not to automate.
- Worked on ML training workflows, RL sandboxes, container networking (CNI), or: checkpoint storage at scale. Direct GPU-cluster experience is a plus.
- Written software in Python or Go. You would rather extend a solid open-source project than rebuild it.
- Worked with researchers, ML engineers, or another team whose goals were not identical to yours.
