Products are the home for your AI Capabilities — your Flows and Agents — and the Surfaces that make them available to the world. Think of a Product as the bridge between what you’ve built and how people use it.
A Product bundles related AI Capabilities together and deploys them through one or more Surfaces. Instead of deploying each Flow or Agent individually, you group them into a Product and choose how users interact with them.
Every Product has two key parts:
Each Surface has its own status (draft, active, or paused) and environment (development or production), so you can test safely before going live.
Maya runs support for a SaaS company and wants to automate common workflows without losing the human touch on complex issues. She creates a Customer Support Product with three Capabilities:
Her customers reach out through different channels, so she adds Surfaces to the same Product:
She sets each Surface to active in the production environment, and her team starts seeing results right away — faster response times and fewer repetitive tickets in the queue. Later, she adds a sentiment analysis Capability to flag frustrated customers for immediate escalation, all within the same Product.
A single Flow or Agent can be added as a Capability in multiple Products. For example, a “summarize text” Flow might be part of your Customer Support Product, your Content Creation Product, and your Research Assistant Product.