Designing agent experience: A practical guide for the era of AX

Agent Experience (AX) is a new design paradigm introduced in 2025 that focuses on enabling AI agents to operate autonomously and effectively, bridging the gap between user experience (UX) and developer experience (DX). As AI agents become more integral to business operations, AX emphasizes the importance of clear tools, documentation, permissions, and fallback paths to ensure agents can function without human intervention. The core principles of AX involve designing systems and workflows that allow agents to understand user interactions predictably and interface with APIs and other infrastructure reliably. AX principles are divided into user-facing, which prioritize predictability, transparency, and coherence for agents interacting with humans, and system-facing, which focus on toolability, recoverability, and traceability for agents integrating with digital systems. Practical implementation of AX includes crafting precise prompts, creating agent-specific documentation, managing API keys, and controlling access through tool curation and role-based permissions. By adopting AX, businesses can ensure their agents act autonomously while maintaining security and reliability, ultimately enhancing the efficiency and satisfaction of both users and developers.