AI Video Creation Orchestrator

Rnb Consultancy; San Francisco (CA); On-site; USD 200,000 - 300,000; Full time

Job summary

Rnb Consultancy in San Francisco is seeking an experienced engineer to design the agent harness behind multi-step video creation workflows, from planning assets to generation, revision and delivery.

You will balance AI capabilities with creator workflow, build observability, and collaborate with product and engineering to turn feedback into better processes, aiming for reliability and usefulness of generated outputs.

Qualifications

• Experience shipping production software for AI-assisted or other complex, multi-step products.
• Hands-on experience building agentic applications or orchestration layers that use tools, state and evaluation—not only standalone prompts or demos.
• Ability to design, debug and operate reliable software services and integrations.
• Sound engineering judgment and a collaborative approach to ambiguous product problems.

Responsibilities

• Design and improve the agent harness behind multi-step video creation workflows, from planning and asset selection to generation, revision and delivery.
• Build dependable connections between agents, creative tools and model providers; make failures observable and recoverable.
• Manage the context, state and tool permissions needed for longer-running creative tasks.
• Define quality checks and evaluation loops for reliability, controllability and the usefulness of generated outputs.
• Work with product, design and engineering partners to turn creator feedback into better workflows, then measure what changes.

Skills

Production software

Agentic apps

Systems integrations

Collaborative mindset

Job description

Rnb Consultancy in San Francisco is seeking an experienced engineer to design the agent harness behind multi-step video creation workflows, from planning assets to generation, revision and delivery.

You will balance AI capabilities with creator workflow, build observability, and collaborate with product and engineering to turn feedback into better processes, aiming for reliability and usefulness of generated outputs.