ByteDance

2 days ago

Machine Learning Engineer - Orchestration

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 38 minutes agoStatus: Live

Job Summary

Build and optimize distributed orchestration and scheduling for large-scale machine learning training and online inference, improving cluster efficiency and resource utilization across Kubernetes/Godel and multi-region/multi-cloud scenarios. Develop next-generation training and online inference architectures for ultra-large-scale recommendation models, extend autoscaling and parallelization, and integrate with platform tooling to improve diagnostics and usability. Partner on research-driven model paradigms and MLops workflows.

Location: San Jose

Workplace: Onsite

Employment Type: Full time

Job Function: Data Science & Machine Learning

Seniority: Mid level

Key Responsibilities

•Optimize resource efficiency in distributed orchestration and scheduling by developing/using scheduling frameworks around the Kubernetes/Godel ecosystem and improving cluster utilization strategies.
•Extend autoscaling and parallelization for models and business operations using load modeling to optimize resource requests and utilization at scale.
•Implement service preemption/eviction and handle borrowing/mixed deployment across resources, clusters, and scenarios spanning multiple data centers, regions, and clouds.
•Build distributed training architecture for next-generation ultra-large recommendation models, including robust training runtime, GPU synchronization, and distributed computing APIs for research paradigms.
•Construct online orchestration architecture for next-generation recommender systems by optimizing distributed inference around online training scenarios and integrating research/experimental models with MLops.

Key Requirements

•Bachelor’s degree or above in Computer Science or a similar field.
•At least 5 years of experience with strong hands-on coding in Go or Python in a Linux environment.
•Familiar with open-source distributed scheduling frameworks such as Kubernetes (K8S), Yarn, Flink, MapReduce, Mesos, and Celery, with experience building machine learning systems.
•Strong understanding of distributed systems and experience designing, developing, and maintaining large-scale distributed systems.
•Excellent logical analysis and ability to abstract and split business logic, with good documentation habits.

Experience:Machine learningDistributed systemsRecommendation

Education:Bachelor's

Skills:Logical analysisCommunicationSelf-motivationDocumentation

Tech Stack:GoPythonLinuxKubernetesGodelYarnFlinkMapReduceHadoopMesosCeleryPyTorchTensorFlowK8SRLFinetuneDistillationGPUAutoScaling

Company Brief

ByteDance

Develops consumer internet and content platforms, including TikTok and other apps for short-form video, news, and entertainment. It also builds advertising, commerce, and creator tools that connect audiences, brands, and publishers across global markets.

Industry: Digital Media

Company Size: Enterprise (1,001+ employees)

Revenue: USD 1B+

Growth: Established Company

Headquarters: Beijing, China

Founded: 2012

Website LinkedIn

Loading job details...

Preparing the role view and application actions.

ByteDance

2 days ago

Machine Learning Engineer - Orchestration

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 38 minutes agoStatus: Live

Job Summary

Location: San Jose

Workplace: Onsite

Employment Type: Full time

Job Function: Data Science & Machine Learning

Seniority: Mid level

Key Responsibilities

•Optimize resource efficiency in distributed orchestration and scheduling by developing/using scheduling frameworks around the Kubernetes/Godel ecosystem and improving cluster utilization strategies.
•Extend autoscaling and parallelization for models and business operations using load modeling to optimize resource requests and utilization at scale.
•Implement service preemption/eviction and handle borrowing/mixed deployment across resources, clusters, and scenarios spanning multiple data centers, regions, and clouds.
•Build distributed training architecture for next-generation ultra-large recommendation models, including robust training runtime, GPU synchronization, and distributed computing APIs for research paradigms.
•Construct online orchestration architecture for next-generation recommender systems by optimizing distributed inference around online training scenarios and integrating research/experimental models with MLops.

Key Requirements

•Bachelor’s degree or above in Computer Science or a similar field.
•At least 5 years of experience with strong hands-on coding in Go or Python in a Linux environment.
•Familiar with open-source distributed scheduling frameworks such as Kubernetes (K8S), Yarn, Flink, MapReduce, Mesos, and Celery, with experience building machine learning systems.
•Strong understanding of distributed systems and experience designing, developing, and maintaining large-scale distributed systems.
•Excellent logical analysis and ability to abstract and split business logic, with good documentation habits.

Experience:Machine learningDistributed systemsRecommendation

Education:Bachelor's

Skills:Logical analysisCommunicationSelf-motivationDocumentation

Tech Stack:GoPythonLinuxKubernetesGodelYarnFlinkMapReduceHadoopMesosCeleryPyTorchTensorFlowK8SRLFinetuneDistillationGPUAutoScaling

Company Brief

ByteDance

Industry: Digital Media

Company Size: Enterprise (1,001+ employees)

Revenue: USD 1B+

Growth: Established Company

Headquarters: Beijing, China

Founded: 2012

Website LinkedIn