Senior Engineer

此職位由香港理工大學 (PolyU) 發布。 職位編號：261109004-IE 學系/部門：University Centre for AI Computing 類別：管理/專業職位 截止日期：16 November 2026 來源：香港理工大學招聘網 申請及詳情：https://jobs.polyu.edu.hk/job_detail.php?job=261109004 ---

(Ref. 261109004-IE)

Duties

The appointee will be required to:

(a) lead the design and development of the large language model training platform, build capabilities for GPU resource management, training job scheduling, inference services and MLOps platform to support the efficient training and iteration of large language models;

(b) take charge of the construction and optimisation of GPU cluster, and build the computing platform based on Kubernetes + NVIDIA GPU Operator, including resource scheduling, containerised environment, and the configuration and optimisation of GPU software stack (CUDA, Driver, NCCL, etc.);

(c) develop the core toolchain for the training platform, training job orchestration, automated pipelines, base image management, data loading and model version management;

(d) collaborate with algorithm teams to optimise distributed training frameworks, such as PyTorch Distributed, Megatron and SGLang, so as to improve training efficiency and stability;

(e) establish platform monitoring and automated operation and maintenance systems, and enhance monitoring, alerting and fault diagnosis capabilities for GPU, network, storage and training jobs; and

(f) perform any other duties as assigned by the Head of Unit or his/her delegates.

Qualifications

Applicants should have:

(a) a master’s degree in Computer Science, Telecommunications, Electronic Engineering or a related discipline;

(b) at least five years of solid experience in system backend development, with some experience in distributed systems or large-scale computing platform development;

(c) familiarity with large language model training workflows, with understanding of the full model lifecycle, including training, inference and evaluation;

(d) proficiency in programming languages such as Python / Go, with solid software engineering capabilities;

(e) hands-on experience in Kubernetes, Docker, GPU scheduling and AI computing environment setup;

(f) familiarity with high-performance computing technologies, including NVIDIA GPU architecture, CUDA, NCCL and InfiniBand; and

(g) fluency in Cantonese and English.

Preference will be given to those with:

(a) experience in building thousand-GPU training clusters;

(b) familiarity with large language model training and inference frameworks, including Ray, DeepSpeed, PyTorch, Megatron, vLLM and SGLang;

(c) experience in developing GPU training platforms, MLOps platforms or distributed systems; and

(d) familiarity with large-scale storage systems, such as HDFS, GPFS and JuiceFS.

Conditions of Service

A highly competitive remuneration package will be offered. Initial appointment will be on a fixed-term gratuity-bearing contract. Re-engagement thereafter is subject to mutual agreement.

Consideration of applications will commence on 16 November 2026 until the position is filled.