Staff ML Systems Engineer: Scalable AI Infrastructure

Meta; Sunnyvale (CA); On-site; USD 183,997 - 257,000; Full time

Job summary

Meta is seeking a Staff Software Engineer to join the Systems ML Engineering team, focusing on building and scaling the infrastructure and software systems that power large-scale machine learning workloads across Meta's production fleet.

In this role, you will architect and own critical components of the ML systems stack, spanning training infrastructure, model serving, distributed computing frameworks, and ML platform tooling.

Qualifications

• 8+ years of software engineering experience focusing on systems or ML infrastructure.
• Experience designing and implementing large-scale distributed systems (training orchestration, serving, or data pipelines).
• Experience with performance analysis and optimization of compute-intensive workloads (profiling, benchmarking, bottleneck ID).
• Experience leading end-to-end delivery of complex technical projects across teams.
• Proficiency in C++ or Python for production ML or infrastructure systems.

Responsibilities

• Design scalable ML infrastructure components (training frameworks, model serving, ML platform tooling).
• Lead architectural decisions for major ML infrastructure initiatives with trade-off analysis.
• Identify and resolve performance bottlenecks in distributed ML systems through profiling and optimization.
• Define SLAs for ML services, build dashboards, alerts, and runbooks to reduce MTTR.
• Collaborate with researchers, engineers, and infra teams to translate model requirements into production systems.

Skills

Distributed systems design

Performance analysis

End-to-end project leadership

C++

Python

Job description

Meta is seeking a Staff Software Engineer to join the Systems ML Engineering team, focusing on building and scaling the infrastructure and software systems that power large-scale machine learning workloads across Meta's production fleet.

In this role, you will architect and own critical components of the ML systems stack, spanning training infrastructure, model serving, distributed computing frameworks, and ML platform tooling.