Role Overview
Cohere is a security-first enterprise AI company building foundation models and products. The Internal Infrastructure team builds infrastructure and tools to train, evaluate, and serve models. You'll own the developer platform researchers use for writing, building, and debugging training code: dev environments, container images, CI, and training instrumentation. Includes a compensated 24x7 on-call rotation.
Responsibilities
- Run secure developer environments that stay controlled and accelerate developer velocity as the research org scales.
- Own container image build and CI/CD tooling.
- Instrument training code and build self-service tooling so researchers can debug failed runs themselves.
- Set infrastructure best practices through AI enablement, documentation, and office hours.
- Troubleshoot production failures and performance problems, leaving systems easier to operate.
- Build automation, infrastructure-as-code, and observability to keep the platform maintainable.
Requirements
- Run a developer platform in production: CI, image builds, dev environments, or internal tooling many engineers depend on.
- Made ML engineers or researchers faster without making the platform fragile.
- Strong Linux and Kubernetes skills to debug production systems; GPU cluster experience a plus.
- Proficiency in Python or Go; prefer extending solid open-source projects over rebuilding.
- Experience collaborating with researchers, ML engineers, or teams with differing goals.
- Proven ability to find bottlenecks, propose fixes, and drive them in fast-moving environments.