SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Intercom is seeking a Staff AI Designer to shape the behavior, capabilities, and integration of AI systems within their customer support platform. This role sits at the intersection of product design, machine learning, and user experience—going beyond traditional interface design to architect how AI agents operate, make decisions, and integrate into real-world customer workflows.
Key responsibilities include:
**Define AI Capabilities**: Identify high-value opportunities for AI across the product. Determine what AI should and should not do, setting clear boundaries between autonomous operation and human control. Make strategic calls on scope and limitations.
**Design AI Behavior & Orchestration**: Architect AI-driven workflows and agentic systems that handle multi-step interactions, agent-to-agent communication, and real-world scenarios over time. Establish decision frameworks for when systems should act, ask for input, escalate, or defer to humans. Design for uncertainty, failures, and edge cases as core system properties.
**Shape Human-AI Workflows**: Design end-to-end experiences where users manage and interact with AI systems. Define how AI integrates into actual user workflows, not isolated interactions. Determine where humans guide, review, or override AI decisions. Ensure users can understand, trust, and recover from failures.
**Define Quality & Reliability**: Establish what "good" looks like (trust, accuracy, effort reduction, usability). Design evaluation scenarios and feedback loops. Analyze system outputs to identify failure modes. Drive reliability at the system level to prevent agentic breakdowns.
**Cross-functional Collaboration**: Work closely with ML scientists, product managers, and product designers. Influence direction from zero-to-one problem stages. Operate autonomously in a fast-moving environment. Prototype and run focused experiments. Stay close to system behavior by regularly reviewing outputs and raising quality standards.
This is a highly specialized role for someone who understands LLM capabilities and limitations, has experience defining design principles for AI systems, and can operate through ambiguity while maintaining high quality standards.
