Intern, AI Systems Performance Engineer
Samsung Semiconductor2 days ago
San Jose, CA, USAIntern
Responsibilities
- Build runtime adapters and a trace record/replay harness for AI agent frameworks.
- Develop micro-kernel and end-to-end benchmarks stressing CPU, memory, storage, and accelerator subsystems.
- Profile agent workloads using perf, eBPF, hardware PMUs, and NVML/DCGM.
- Measure DDR5, GPU HBM, and CXL.mem bandwidth and latency across Intel Xeon and AMD EPYC servers.
- Evaluate multi-agent KV-cache sharing and CXL memory expansion using Samsung CXL expanders.
- Build an initial HBM/DDR5/CXL memory-tiering cost model for long-running MoE agent workloads.
- Present weekly findings and deliver a final report and presentation.
Requirements
- Pursuing a PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
- Strong programming skills in Python and C/C++ in a Linux environment.
- Solid understanding of computer architecture, including CPU caches, memory hierarchy, and NUMA.
- Experience with performance profiling tools such as perf, eBPF, VTune, or Nsight.
- Preferred experience with LLM serving frameworks such as vLLM or SGLang, or with GPU programming.
- Preferred familiarity with HBM, CXL, or heterogeneous memory systems.
- Preferred experience with AI agent frameworks or benchmark development.
Benefits
- Hourly base pay of $28—$68 USD.
- Daily onsite work at the San Jose headquarters under the company’s Flexible Work policy.
- Relocation support for employees living 50 or more miles from the new office.
- Monthly housing stipend for the summer internship.
- Medical, dental, and vision benefits, charitable giving match, emotional wellness support, onsite café and gym, virtual fitness classes, and a flexible work environment.