Senior Software Engineer - AI Frameworks
Core
Architect and implement efficient tensor computation primitives and software abstractions for custom AI accelerators, extending PyTorch features for model onboarding, optimization, and execution.
Role type
Senior IC software engineer (AI frameworks & accelerators)
Builds
High-performance inference and training stacks for custom AI accelerators (NPUs, GPUs)
Domain
AI/ML infrastructure, LLM inference and training
Deliverable
production ML models
Required skills
C, C++, Python, tensor computation primitives, software abstractions, PyTorch internals, high-performance kernel development, profiling, optimization
Preferred skills
vLLM, SGLang, CUDA, Triton, quantization (PTQ/QAT), distributed parallelism (TP, PP, DP)
Technologies
PyTorch, vLLM, SGLang, CUDA, Triton, NPUs, GPUs
Responsibilities
Design and develop high-performance kernels for NPUs and GPUs; optimize AI inference stacks including scheduling and KV cache management; extend PyTorch features for model execution; define requirements and deliver practical solutions to technical challenges
Seniority
Senior, hands-on IC
