R&DRemote, GlobalFull-time

AI Agent Orchestrator


AI-native

Birdhouse is an AI-native agency.

All work is executed by AI agents - people focus on supervision, direction, and quality control. If you're excited about working at the frontier of AI driven development, you'll fit right in.

As an AI Agent Orchestrator, you own the systems and workflows that turn a business goal into coordinated agent work and shipped output. You design how agents are tasked, how they hand off to one another, and how their results are checked before they reach a client.

What you'll do

  • Decompose product and delivery goals into agent-executable tasks, and design the orchestration flows that route them across specialized agents.

  • Build, tune, and maintain multi-agent pipelines (prompts, tools, guardrails, and hand-off logic) so work moves reliably from brief to finished deliverable.

  • Supervise agent runs in real time: catch drift, unblock stuck agents, and intervene when human judgment is needed.

  • Define quality gates and review checkpoints, and measure agent output against them so quality is enforced, not hoped for.

  • Turn recurring failure patterns into reusable playbooks, evals, and improvements to the orchestration layer.

  • Partner with clients and internal leads to translate ambiguous requirements into clear, executable direction for agents.

What we're looking for

  • Strong systems thinking. You can hold a multi-step workflow in your head and see where it will break.

  • Hands-on experience directing LLM-based agents or automation pipelines, and a feel for prompt design, tool use, and context management.

  • Sharp quality instincts. You can look at generated output and tell what's good, what's wrong, and why.

  • Excellent written communication. You specify work precisely, because vague direction produces vague results.

  • Comfort working async and iterating fast in a frontier, still-being-invented discipline.

Nice to have

  • Experience with agent frameworks, orchestration tools, or building evals for LLM systems.

  • A background in engineering, product, or delivery ops where you owned outcomes end-to-end.

  • Prior work at an agency, startup, or any environment where you shipped under real deadlines.

We care more about what you've shipped and how you think than where you went to school. If you're excited to work at the frontier of AI-driven delivery, directing agents rather than doing the rote work yourself, you'll fit right in.

Apply for this role

We review every application personally and will be in touch.

PDF, max 5 MB.