- San Jose, California, United States
- On-site
- Internship · Intern
- $28–$68 per hour
- 3+ years
Research AI system architecture up to rack level, focusing on memory hierarchies for long-horizon agents.
Compensation
$28—$68 USD
Responsibilities
- Characterize LLM and agentic inference workloads
- Research memory hierarchies for AI inference
- Propose new architectures and system optimizations
- Build and extend performance models and simulators
- Measure accuracy impact of memory optimizations
Requirements
- PhD student currently pursuing degree in Computer Science
- Nice to have: first-author publications
- Nice to have: top-tier computer architecture venues
- Nice to have: machine learning venues
- Nice to have: vLLM