Teams at Together AI

Design and operate multi-petabyte, high-performance storage for large-scale AI/HPC clusters. Architect parallel filesystems and object stores, build Kubernetes-native storage operators, optimize RDMA/InfiniBand and NVMe-based data paths for 10–50 GB/s per node, implement multi-tier caching and prefetching, ensure observability and 99.9%+ uptime, and collaborate with ML/SRE teams while driving cost and performance improvements.

As an AI Researcher, you will develop novel architectures and optimizations in machine learning, collaborating with teams to enhance open models and ensuring cutting-edge research practices.

As a Forward Deployed Engineer, you will optimize inference engines and guide customers in deploying AI models effectively, ensuring high-value outcomes and driving product improvements.