Principal Agentic Ai Engineer

Zetaglobal

New York, United States
On-site
Agentic ai workflows
Rag + tool-using workflows
Ace (observe → reason → act → evaluate)
Zeta Global is seeking a hands-on Agentic AI engineer to build bidder-adjacent agents that improve campaign performance using RAG + tool-using workflows in a closed-loop ACE framework

Job Summary

  • Zeta Global is seeking a hands-on Agentic AI engineer to build bidder-adjacent agents that improve campaign performance using RAG + tool-using workflows in a closed-loop ACE framework.
  • The role involves building agentic workflows for campaign control actions, implementing production-grade RAG, creating safe tool/action interfaces, and owning AgentOps including evaluation harnesses and online experimentation.
  • The engineer will partner with various teams to define the boundary between deterministic bidding and agent-driven control-plane decisions, while adding end-to-end observability and reliability patterns.

Matching Summary

Zeta Global is seeking a hands-on Agentic AI engineer to build bidder-adjacent agents that improve campaign performance using RAG + tool-using workflows in a closed-loop ACE framework.

Skills & Requirements

Must-have

  • Agentic AI workflows
  • RAG + tool-using workflows
  • ACE (Observe → Reason → Act → Evaluate)
  • Low-latency AdTech environment
  • Production-grade RAG implementation
  • Safe tool/action interfaces
  • AgentOps (eval harnesses, regression, A/B testing)

Nice-to-have

  • Sophisticated marketing simplification
  • Personalized consumer experiences
  • Cross-functional partnership

Key Requirements

  • Hands-on Agentic AI engineering experience
  • Experience with RAG and tool-using workflows
  • Experience with ACE (Observe → Reason → Act → Evaluate) frameworks
  • Experience in low-latency AdTech environments
  • Experience with AgentOps and evaluation harnesses
  • Experience with online experimentation (A/B testing)

Work Rights

Not specified

Tailored Resume

Cover Letter