AI Infrastructure Engineer – Training Framework Optimization

MillenniumHong KongOn-siteFull-timeMid level, 2–5 yearsListed 6 hours ago

Apply now

About this role

AI Infrastructure Engineer – Training Framework Optimization

Job Description: AI Infrastructure Engineer – Training Framework Optimization

Job Function Summary

We are seeking an AI Infrastructure Engineer to optimize large-scale model training efficiency and high-throughput data pipelines. You will profile GPU bottlenecks, optimize multi-GPU distributed communication, and build scalable data loaders to support our deep learning research at scale.

Principal Responsibilities

·         Optimize end-to-end model training throughput and hardware utilization across large-scale GPU clusters.

·         Maintain and optimize high-performance, low-latency Data Loaders.

Qualifications/Skills Required

·         Strong Python programming skills.

·         Deep familiarity with PyTorch internals.

·         Solid understanding of GPU architecture, including CUDA kernels, streams, events, and contexts.

·         Proficient in using profiling tools to identify and resolve model compute, memory, and communication bottlenecks.

·         Strong understanding of multi-GPU distributed communication principles and optimization techniques.

·         Familiarity with static graph compilation technologies (e.g., torch.compile, XLA, TensorRT).

·         Ability to architect and scale high-throughput Data Loaders utilizing asynchronous I/O and concurrency techniques.

Preferred Qualifications

·         Familiarity with common model architectures (such as Attention mechanisms) and the ability to collaborate with researchers on co-designing more efficient model structures.

·         Hands-on experience with mixed-precision training, model quantization, and convergence/accuracy alignment techniques.