Platform Engineer - Enterprise AI

Role overview Build and operate the software and infrastructure behind a platform for deploying AI systems across enterprise environments. You’ll combine backend engineering with infrastructure automation, reliability, and observability, making deployments repeatable even when customer environments are separately controlled.

Responsibilities - Develop and operate platform services and a model gateway. - Build deployment systems for customer cloud accounts, dedicated instances, and sovereign environments. - Create and maintain CI/CD pipelines, infrastructure-as-code, and deployment tooling using Kubernetes, Helm, Terraform or Pulumi, and GitOps practices. - Design for secure multi-tenant operation and clear service ownership. - Establish observability through service-level objectives, structured logs, metrics, traces, and alerts. - Track model and compute usage by tenant, and take part in on-call operations and incident response.

Requirements - Six or more years of platform or infrastructure engineering supporting production SaaS. - Experience shipping software that customers operate in their own environment, such as a private cloud network, self-hosted installation, or air-gapped setup. - Strong production software engineering skills in at least one backend language, such as Go, Ruby, Python, or TypeScript, with ability to work across others. - Deep experience with one cloud provider and working knowledge of another, including networking, key management, identity and access controls, Kubernetes, and infrastructure automation. - Experience carrying an on-call rotation, handling incidents, and operating observability tooling. - Familiarity with Postgres, object storage, queues, and vector stores.

Nice to have - Model serving or gateway experience, including tools such as vLLM, LiteLLM, or Triton. - Experience implementing SOC 2 controls in CI or operating AI systems in regulated, highly controlled environments.

Benefits and work setup Hybrid role with regular office attendance in New York City. The work includes production software development, operational ownership, and deployments across customer-controlled environments.