AI Infrastructure Engineer

Mercor is seeking an AI Infrastructure Engineer to own the accelerator platform end to end, including GPU clusters across AWS, GCP, neoclouds, and on-prem. Responsibilities include building scheduling, distributed training infrastructure, LLM serving, benchmarking, health checks, cost attribution, and internal tools. The role requires 5+ years experience in distributed systems or ML infrastructure, proficiency in Python and Go/C++, deep Kubernetes and container experience, and hands-on GPU work.

5+ years building production distributed systems or ML infrastructure; production-quality Python plus Go or C++; deep hands-on Kubernetes and container experience; working knowledge of Slurm; run GPU workloads at scale; experience with distributed training/serving stacks (PyTorch Distributed, Ray, vLLM, Megatron-LM); deep experience with at least one major cloud provider; observability tools (DataDog, Prometheus, PromQL, Grafana); own projects end to end.