Principal SW Engineer - LLM Serving (Cloud AI)
Core
Developing software solutions for Inference Acceleration for LLMs, including serving frameworks and optimizing neural networks on multicore systems.
Role type
Principal Software Engineer (LLM Serving)
Builds
Commercial software for inference acceleration and LLM serving
Domain
Cloud AI, Machine Learning, Systems Engineering
Deliverable
production ML models
Required skills
LLM serving frameworks (vLLM), PyTorch, C++, Python, neural network optimization, multicore system software development, SoC architecture understanding, performance modeling
Preferred skills
Machine learning accelerators, neural network operators, linear algebra, multi-modal and reasoning models
Responsibilities
Design, compile, model performance, and analyze bottlenecks for inference acceleration software; execute and optimize neural networks on multicore architectures; manage large commercial software projects from R&D to deployment
Seniority
Principal, hands-on IC
