Principal Engineer, AI Serving Framework Architect (Software)

Samsung Semiconductor is seeking a Principal AI System Architect to lead research on dynamic scheduling and optimization for AI inference at scale. The role involves working on memory-centric systems, RAG vector DB acceleration, and KVCache placement strategies.

• Lead research teams in Korea and propose technical direction.
• Research dynamic scheduling methodologies for maximizing AI inference performance in multi-rack scale memory-centric systems.
• Investigate methods to accelerate search operations in RAG's vector DB and AI Agent's knowledge-graph using compute-capable memory.
• Study strategies for optimally placing KVCache and vector DB in hierarchical memory to minimize SSD accesses.
• Propose software designs for implementing optimization algorithms on platforms like vLLM.

• PhD in Computer Science or related field with 10+ years experience in AI Serving Framework.
• Led a project to build and optimize LLM Inference Software Stack on multi-rack scale system.
• Extensive experience in designing AI Inference Software Stacks for heterogeneous devices.
• In-depth understanding of vLLM internal architecture and operation.
• Proficiency in AI Inference System Profiling and optimization.