Agent Builder - AI Systems
Core
Training advanced AI agents by providing high-quality human feedback to shape proactive, multi-step LLM systems capable of operating as autonomous agents.
Role type
Senior AI Systems Engineer (Human-in-the-loop Agent Training)
Builds
Proactive, multi-step LLM systems and complex real-world architectural workflows for autonomous agents
Domain
Artificial Intelligence / Large Language Models / Agent Orchestration
Deliverable
production ML models
Required skills
Backend engineering, AI automation, complex systems integration, multi-stage coordination workflow design, live non-mocked environment development, multi-turn system interaction management, API integration, persistent state and session tracking, privacy leak detection, authority escalation prevention, indirect prompt injection mitigation
Preferred skills
Experience with Supabase, Gmail, and other live tools, designing multi-stage coordination workflows where information gathering leads to reasoned outcomes
Technologies
Java, JavaScript, Python, SQL, LLM
Responsibilities
Train advanced AI agents by providing high-quality human feedback, help shape proactive multi-step LLM systems, evaluate system behavior in live environments, support development of orchestration and automation capabilities, analyze multi-turn interactions to surface actionable feedback, identify and report nuanced failure modes in agent behavior
Seniority
Mid-level to Senior, hands-on IC