Evaluate AI agents with software engineering and technical feedback skills. Outlier offers remote freelance work for contributors in listed countries.
This work supports leading AI organizations training large language models as proactive, multi-step agents. You will assess complex system behaviors and help improve real-world architectural workflows. The work suits developers with at least two years in backend engineering, AI automation, or complex systems integration, plus experience with live environments.
Key Responsibilities: Debug agent traces across multi-turn interactions
Write detailed evaluation rubrics for agent performance
Stress test autonomous agents against complex workflows
Identify subtle failures in system behavior
Requirements: At least two years of relevant skilled experience
Maintain modular, production-grade software systems
Work effectively in live, non-mocked environments
Give high-density technical feedback on complex behaviors
Experience with backend engineering or AI automation
Use software engineering judgment, agent debugging, and evaluation skills to strengthen autonomous AI systems through Outlier projects.

#J-18808-Ljbffr

📌 AI Agent Evaluation Software Engineer (New Brunswick)
🏢 Outlier
📍 New Brunswick