Software Engineer 5 – Agent Platform, AI Platform

Overview

Build the foundational platform for Netflix's production AI agents as a senior individual contributor. Own core agent primitives, improve agent quality and safety, and guide technical direction across partner teams.

What you'll do

• Design, build, and operate a core agent primitive end to end.
• Capture agent trajectories and convert outcomes into evaluation signals.
• Build memory and feedback loops that improve agent behavior across runs.
• Develop tracing, quality signals, and guardrail capabilities.
• Define architectures and interfaces for shared platform foundations.
• Partner with ML, product, security, compute, and data teams.
• Iterate with agent developers to turn usage feedback into platform capabilities.

What you'll need

• Typically 8+ years of significant software engineering experience.
• Experience shipping and operating production systems.
• Hands-on experience deploying and operating LLM agents in production.
• Experience building agent evaluation suites, tracing, and quality signals.
• Strong distributed-systems design and operations fundamentals.
• Proficiency in Python and its packaging tooling.
• Proficiency in Java, Go, C/C++, Rust, or Zig.

Nice to have

• Experience building high-scale memory or state systems for agents.
• Experience with agent identity, permissioning, safety, guardrails, or cost governance.
• Credibility in a relevant open-source ecosystem.
• Familiarity with Temporal, FastAPI, PostgreSQL, Kubernetes, and CI/CD practices.

Details

• Remote role in the USA.
• Annual salary range: $466,000–$750,000.
• Compensation varies based on location.