ML Platform & Orchestration Engineer

Bytedance; San Jose, Northern (CA, KY); Hybrid; USD 140,000 - 210,000; Full time

Job summary

ByteDance seeks a PhD-level engineer to design and build scalable orchestration for ML platforms, including Kubernetes Operators and lifecycle management for jobs and services.

You will improve GPU utilization, implement multi-cluster serving, and advance autoscaling and disaster recovery across disaggregated clusters. A strong background in distributed systems and open-source tooling is valued.

Qualifications

• PhD in Computer Science or related field.
• Strong knowledge of Go, C++, or Python.
• Experience with Linux, OS concepts, networks, concurrency, and distributed systems.
• Ability to analyze systems via metrics, logs, and profiling.
• Proven ownership and collaboration through projects.

Responsibilities

• Design and build foundational orchestration capabilities for ML platforms.
• Develop multi-tenant resource and quota systems and improve GPU utilization.
• Create lifecycle orchestration for model serving and multi-cluster operations.
• Implement serving and traffic management for disaggregated clusters.

Skills

Go

C++

Python

Distributed systems

Data structures & algorithms

Education

PhD in Computer Science / Software Engineering

Tools

Kubernetes

Container runtimes

Volcano

Koordinator

OpenKruise

KServe

Job description

ByteDance seeks a PhD-level engineer to design and build scalable orchestration for ML platforms, including Kubernetes Operators and lifecycle management for jobs and services.

You will improve GPU utilization, implement multi-cluster serving, and advance autoscaling and disaster recovery across disaggregated clusters. A strong background in distributed systems and open-source tooling is valued.