Remote job
AI Infrastructure Engineer [Tooling]
Job details
About this role
Role overview
Build and operate the software and infrastructure that moves AI agents from development into customer cloud environments. This hands-on individual contributor role spans developer tooling, Kubernetes operations, model serving, and production reliability. The role can be calibrated at mid or senior level, with broader technical direction and mentoring expected at the senior level.
Responsibilities
- Develop Python or Go tools, SDKs, command-line interfaces, and services for deploying, testing, and debugging agent workloads. - Run and tune LLM inference systems, addressing GPU scheduling, autoscaling, batching, latency, and cost. - Build agent runtime capabilities such as tool execution, sandboxing, state management, and safe version rollouts. - Own model and prompt versioning, evaluation workflows, artifact management, and releases from staging to production. - Create Kubernetes operators and controllers, and operate infrastructure provisioned with Terraform across cloud environments. - Improve observability, security, CI/CD, GitOps, and self-service automation in partnership with AI, product, and platform teams.
Requirements
- At least five years in infrastructure, platform, or backend engineering; seven or more for senior level. - Strong Python or Go engineering skills and experience shipping tools or services used by other engineers. - Production experience serving LLM workloads and building infrastructure for AI agents or LLM applications. - Practical MLOps experience, including model or prompt versioning, evaluation and release pipelines, and observability. - Hands-on Kubernetes experience, including operators or controllers, plus Terraform and at least one major cloud provider. - Experience owning CI/CD and GitOps-based releases, solid networking fundamentals, and operating production systems.
Nice to have
- GPU infrastructure, agent frameworks or protocols, LLM observability, GitOps controllers, or cloud security and compliance automation. - Deep networking expertise or open-source contributions in AI infrastructure or Kubernetes.