NEAR AI is hiring a LLM Inference Engineer

NEAR AI is hiring a remote LLM Inference Engineer to push how large language models are served. The team builds decentralized, confidential ML infrastructure so that open-source AI can run at global scale without handing user data to a single provider.

What you'll do

• Architect and maintain production, high-traffic LLM serving systems
• Optimize throughput, latency, and cost for leading open-source models
• Work close to the metal on GPU performance across the serving stack

Requirements

• Hands-on LLM inference experience, including debugging and optimizing engines such as SGLang, vLLM, or TensorRT
• Deep knowledge of modern GPU architectures and how to exploit them with PyTorch, Triton, CuTe, or CUDA
• A track record of designing and running end-to-end, high-traffic LLM serving systems
• Nice to have: trusted execution environments (TEE), or open-source contributions to an inference engine

San Francisco or remote. Full details and application at the link.