Intern - AI Systems and Infrastructure Engineering
Core
Develop and optimize systems software, profiling tools, and experimentation frameworks for Large Language Models (LLMs) and Agentic AI workloads across GPU platforms and heterogeneous memory systems.
Role type
AI Systems Software Engineering Intern
Builds
Systems software, profiling tools, and experimentation frameworks for LLM training, inference, and Agentic AI
Domain
Semiconductor industry + AI infrastructure (memory, storage, interconnects)
Deliverable
production ML models | infrastructure
Required skills
Python, C/C++, Linux development, GPU performance analysis, LLM architecture (transformers, KV cache, attention), AI frameworks (PyTorch, vLLM, TensorRT-LLM)
Preferred skills
LLM runtime optimization, KV-cache/memory management implementation, GPU optimization (CUDA, Triton, NCCL, RDMA), heterogeneous memory architectures (HBM, CXL, NVMe)
Responsibilities
Develop profiling tools and experimentation frameworks for AI workloads; Design and evaluate memory/state-management techniques (caching, tiering, compression); Characterize and optimize AI workload execution across GPUs, CPUs, and distributed infrastructure; Build benchmarking and automation capabilities for data placement and scheduling; Collaborate on representative AI workloads and technical publications
Seniority
Intern