Intern, AI Application Engineer

About the role

Internship with the SoC Architect team: deploy and benchmark LLM inference (vLLM) and KV cache offloading to network-attached SSD storage. Onsite; posting lists Folsom CA and San Jose CA HQ.

Requirements

• Pursuing Bachelors, Masters or PhD in CS, CE, EE or related
• At least 1 academic quarter/semester remaining
• Hands-on LLM inference with Python and PyTorch on NVIDIA GPUs under Linux
• Understanding of LLM serving internals
• Benchmarking and profiling experience

What you'll do

• Deploy vLLM serving Llama-3.3-70B on 4x H200 server
• Build automated benchmark harness
• Configure KV cache offloading to NVMe-oF storage
• Run experiment sweeps
• Analyze bottlenecks with Python

Where this was posted

• greenhouse.com

  last seen 20m ago
  View on greenhouse.com