AI Software Architect
Core
Design, develop, and optimize software features for AI inference and training frameworks (PyTorch, vLLM, SGLang) to enhance deep learning capabilities on Intel AI accelerators and GPUs.
Role type
Senior AI Software Architect
Builds
AI inference serving frameworks and deep learning training/fine-tuning capabilities for Intel AI accelerators and next-generation GPUs
Domain
Data Center / AI Infrastructure / Deep Learning
Deliverable
production ML models
Required skills
Python, C++14/17, PyTorch, vLLM, SGLang, Large Language Models (LLMs), computer architecture, hardware-software optimization, complex system debugging
Preferred skills
Deep learning models for image/video generation, kernel development (CUTLASS, Triton), compiler algorithms for heterogeneous systems, Fuser optimizations
Technologies
PyTorch, vLLM, SGLang, CUTLASS, Triton, Intel AI accelerators, GPUs
Responsibilities
Design and develop software features for AI frameworks; Architect and analyze state-of-the-art features across frameworks; Enhance deep learning training and inference capabilities; Optimize software stack performance for deep learning workloads; Participate in open-source community development; Debug complex issues in multi-layered software systems; Conduct performance analysis using profilers; Build technology demonstrators and proof-of-concepts
Seniority
Senior, hands-on IC
