Meta AR/VR Job | Engineering Manager, PyTorch - AI Acceleration
Job（岗位）: Engineering Manager, PyTorch - AI Acceleration
Type（岗位类型）: Engineering
Citys（岗位城市）: Bellevue, WA
Date（发布日期）: 2024-4-29
Summary（岗位介绍）
AI Acceleration is an org within PyTorch. It's responsible for making PyTorch run performantly and reliably on new hardware from external vendors (NVIDIA, AMD, etc.) and Meta's own AI chips.
We are looking for an engineering manager to support internal GPU enablement efforts - using our team's expertise to make GPU inference and training more efficient, reliable, scalable, and simple.
The ideal candidate should have strong technical skills - GPU / ML Systems knowledge is preferred, though not required. We work closely with internal product groups - XFN is a key part of the job.
We have assembled some of the best engineers in the industry to solve some of the world's most foundational problems at a scale that frankly defies comprehension. This is a chance to support these engineers and at the same time learn about industry-leading trends in high-performance computation.
Qualifications（岗位要求）
2+ years of experience in managing a team of HPC/GPU engineers of varied skill levels.
Demonstrated experience recruiting, building, structuring, leading technical organizations, including performance management.
Experience with cross functional collaboration with product ML or AI framework teams.
GPU/CPU optimization skills
Description（岗位职责）
Grow a team of domain experts within AI Acceleration
Communicate, collaborate, and build relationships with clients and peer teams to facilitate cross-functional projects.
Operate strategically and tactically. Develop vision, strategy and help set direction for the team.
Remain up-to-date on ongoing software development activities in the team, help work through technical challenges, and be involved in design decisions.
Additional Requirements（额外要求）
Knowledge of ML frameworks like PyTorch, TensorFlow, ONNX, MXNet, etc.
Experience with different programming models for high-performance computations, e.g. GPU CUDA programming or OpenCL or OpenMP programming.
Experience with ML Systems - GPUs, kernels, compiler, model-specific transformations, quantization, communication optimizations, etc.
