Intern - AI Hardware and Memory Systems
Core
Research and prototype advanced memory and system architecture concepts for next-generation AI accelerators, developing performance models and analyzing AI workloads to optimize system efficiency.
Role type
PhD Research Intern (AI Hardware/Architecture)
Builds
Next-generation memory solutions and AI accelerator architectures
Domain
Semiconductor industry, AI hardware, High-Performance Computing (HPC)
Deliverable
production ML models | product features
Required skills
Computer architecture knowledge, memory system design (HBM, DRAM, caches), performance modeling, architectural simulation, Python/C++ programming
Preferred skills
AI accelerator research, GPU architecture, PyTorch/TensorFlow/CUDA, SystemC, rack-scale system design
Technologies
HBM, DRAM, GPUs, TPUs, SystemC, PyTorch, TensorFlow, CUDA
Responsibilities
Research and evaluate memory architecture concepts for AI accelerators; Develop and analyze performance models for bandwidth, latency, and power; Analyze AI training/inference workloads to identify bottlenecks; Investigate memory hierarchy and interconnect optimizations; Collaborate with cross-functional engineering teams
Seniority
Intern (PhD level)