Machine Learning Ops Engineer, Global SRE

Description

MLOps - Global SRE team is responsible for the stability of machine learning systems under the Global Monetization Products and Technology organization, to ensure the stable and efficient operations of machine learning models from data preparation, development, training, deployment, serving and so on.

Responsibilities
1) Responsible for setting SLOs of online machine learning serving systems, maintaining the stability of the online serving systems.
2) Responsible for maintaining stability of offline machine learning training tasks, improving the success rate of the training tasks.
3) Responsible for rolling out GPU model training in Non-China regions.
4) Responsible for stability of AIGC related machine learning tasks.
5) Responsible for resource management and planning of machine learning resources, including: cost and budget, resource efficiency enhancement, offline and online resources tides, etc.

Requirements

Minimum Qualifications
1) Bachelor's degree in Computer Science or Software Engineering, similar technical field of study, or equivalent practical experience.
2) Expertise in Linux operating systems, networking, storage.
3) Experience programming in at least one of the following programming languages: Python, Go, C, C++, or Java.
4) Experience in troubleshooting application issues, or production operations.
5) Effective communication skills and a sense of ownership and drive.

Preferred qualifications:
1) Experience in SRE of machine learning systems.
2) Experience in SRE of ads/recommendation/search systems.