Core Responsibilities

Design, build, and improve systems that connect AI capability development to production inference. Build and improve model-serving pathways for low-latency, high-throughput inference workloads. Operate and optimize containerized workloads on Kubernetes/GCP, focusing on efficient use of NVIDIA GPUs. Enable shadow serving, canary rollouts, and fast rollback mechanisms for model-backed services.

Requirements

6+ years of production engineering experience. Strong knowledge of model-serving frameworks and runtimes such as vLLM, Triton, or similar systems. Experience with Kubernetes and GCP. Familiarity with benchmarking and evaluation infrastructure.

Additional Information

Experience Level

Senior

Employment Type

permanent

Work Mode

Remote