We are seeking a Member of Technical Staff to join our team working on ClusterMAX™, the industry standard GPU Cloud rating system. We are hiring at all experience levels with competitive compensation.
As part of the interview process, you will complete a paid coding challenge designed to reflect a day in the life on the SemiAnalysis ML Systems team.
Key Responsibilities
- Lead the development of next generation benchmarks and TCO analysis for publication in future versions of ClusterMAX™ and related projects.
- Collaborate with executives and engineers from over 203 neoclouds, hyperscalers, marketplaces, and sovereign projects such as AWS, Azure, GCP, Oracle, CoreWeave, Nebius, Crusoe, Lambda, and Together
- Establish and maintain partnership with AI Chip manufacturers, startups and OEMs such as NVIDIA, AMD, Intel, Google, Amazon, Cerebras, Groq, SuperMicro, Dell, HPE, Lenovo and Cisco
- Build on existing relationships with leading AI labs, investors, startups, and community members to gauge their experience working with cloud providers and contribute expertise
- Author detailed technical research reports analyzing benchmark results, reliability, and ease of use
- Stay current on emerging trends and technologies by attending major conferences such as NeurIPS, MLSys, NVIDIA GTC, OCP, SC, HotChips and more. Travel is encouraged but not required
Qualifications
- Experience with ML frameworks such as PyTorch or JAX
- Experience with GPU or TPU clusters running kubernetes or slurm
- Experience with filesystems such as Weka, VAST, Lustre, and GPFS
- Experience with interconnects such as InfiniBand and RoCEv2
- Experience with ML system benchmarking (GEMMs, nccl-tests, vllm, sglang, mlperf, STAC, HPL, FIO, torchtitan, megatron, etc.)
- Experience working at a hyperscaler, neocloud, server OEM, chip manufacturer, or large scale user of these technologies (preferred)
- Proactive and capable of working in a global, distributed team
