MLOps Engineer

White Circle is an AI Safety company seeking an MLOps Engineer to manage infrastructure for deploying and maintaining production AI models. The role involves building rollout pipelines, monitoring latency and quality, and serving cost optimization using Kubernetes and inference engines.

• Integrate new text and multimodal models into serving paths; Build and maintain rollout pipelines; Create smoke, quality, and performance gates; Operate GPU deployments on Kubernetes; Build dashboards for latency, throughput, GPU usage, etc.; Run A/B and canary rollouts; Debug production issues; Optimize serving cost and reliability.

• Experience with inference serving engines (SGLang, vLLM, etc.); Solid Kubernetes GPU experience; Understanding of multi-node communication and CUDA; Ability to design CI/CD for model serving; Strong observability instincts; Production debugging across Rust to k8s; Clear communication.

• Rust backend experience; Experience with NCCL, UCX, etc.; ClickStack/Datadog; Terraform; DCGM exporter, Prometheus, OpenTelemetry; High model rollout cadence.