About the role#
The Architecture Research Lab is focused on solving memory bandwidth and communication bottlenecks in AI systems. This internship involves researching memory architectures and system optimizations for long-horizon AI agents. You will evaluate these designs using performance models and accuracy studies, with the goal of turning research into publications and future memory product requirements.
What you'll do#
- Characterize LLM and agentic inference workloads to determine memory capacity, bandwidth, and latency requirements.
- Research AI inference memory hierarchies that combine HBM, near-memory compute, High Bandwidth Flash, and host or CXL memory.
- Study how model weights and KV cache are placed, moved, reused, and compressed across memory tiers.
- Propose system optimizations and architectures, ranging from hardware features to runtime and memory-management policies.
- Build and extend performance models and simulators to conduct design-space explorations.
- Measure the accuracy impact of memory optimizations on long-context and agentic benchmarks.
- Collaborate with architects, memory technologists, and system-software researchers to drive projects from definition to results.
- Publish findings at top-tier venues and contribute invention disclosures.
What you'll need#
- Currently pursuing a PhD in Computer Science, with at least 3 years of study preferred.
- At least one academic quarter or semester remaining in your program.
- Strong background in AI system architecture and memory systems, including DRAM/HBM and data movement.
- Deep understanding of LLM and agentic inference, including prefill, decode, KV cache, and long-context workloads.
- Experience with performance analysis and modeling, such as roofline analysis or building simulators.
- Hands-on research experience managing model state across memory tiers, such as KV-cache paging, reuse, offloading, or weight offloading in stacks like vLLM, SGLang, or LMCache.
- Proficiency in Python and C/C++ and experience with PyTorch.
- Preferred: First-author or major contributor status in publications at top-tier computer architecture, systems, or machine learning venues.
Location & details#
- This role is based on-site at our San Jose, California headquarters.
- This is a paid, full-time internship position.
About Samsung Semiconductor
Samsung Semiconductor operates as a subsidiary of Samsung Electronics. The company manufactures semiconductor components for device makers. It started operations in 1974. The organization employs more than 10,000 people.
How to get in at Samsung Semiconductor
Applying to Samsung Semiconductor early matters because recruiters review applications as they arrive. Intern Insider sends an instant alert the moment a role matching your target is published anywhere, so you apply among the first, early applicants get seen before the pile grows. You should aim to be in the first wave of candidates to ensure your resume reaches a human reader. Reaching out to the right person often helps you stand out from the general applicant pool. Intern Insider surfaces the recruiters behind the company's roles so you can reach out directly to ask about the role or a referral, which materially improves response rates. A brief, professional message to a recruiter can clarify your interest and help you learn more about the team.
