Join Baseten as a Distributed Systems Engineer to revolutionize AI performance through cutting-edge GPU networking. Contribute to advanced RDMA implementations and distributed inference optimizations. As part of Baseten's mission, you will work on developing the infrastructure that integrates GPUs into a cohesive system, enhancing the performance of leading AI models.

Your role will focus on networking optimization, applying high-performance protocols like InfiniBand to improve communication and computation efficiency for AI workloads. Key Responsibilities:
Develop RDMA integration to improve speed and bandwidth
Implement networking for Disaggregated KV Cache Offload
Enhance serverless startup times for large models
Measure networking performance on state-of-the-art clusters
Build observability tools for GPU interconnections Requirements:
Expertise in high-performance networking protocols
Proficient in C++ or Python, focusing on hardware interactions
Comfortable analyzing TensorRT-LLM source code
Knowledge of when to utilize custom solutions
Experience with GPUDirect Storage is a plus Be a key player at Baseten, reshaping networking for high-performance AI systems.

📌 Distributed Systems Engineer For Ai Hardware Montreal
🏢 Baseten
📍 Montreal