Own the architecture, implementation, and delivery of a GPU infrastructure platform spanning bare-metal Slurm, Kubernetes, managed inference, observability, and multi-tenant isolation. Lead and line-manage a distributed engineering team, operate GPU and HPC infrastructure, manage vendors and infrastructure partners, establish reliability practices, support internal research and product teams, and complete team hiring. The role is hands-on and requires deep expertise in Linux, Kubernetes, GPUs, networking, storage, and platform operations.

As a Research Engineer Intern, you'll develop state-of-the-art LLM and multimodality models, optimize distributed training frameworks, and collaborate on the application of large models in business scenarios.

Build and maintain the Bare runtime and native modules across platforms. Write C native addons bridging platform APIs into JavaScript, maintain cross-platform build toolchains and CI, debug platform-specific issues, extend runtime core, and write tests. Collaborate with product teams and document work.