Skip to main content

​
What This Video Covers

Agent Studio lets you run multiple AI research tasks in parallel, each inside its own workspace. A workspace acts as a dedicated context area: the first question behaves like a normal AI chat, and later questions can use what has already been asked in that workspace to produce stronger answers. The video walks through opening Agent Studio, creating example workspaces, starting a new agent inside a selected workspace, and sending off separate research questions without waiting on one chat at a time. It shows how agents can keep running in the background while you continue working elsewhere, with notifications available when a task is complete. When an agent finishes, the result looks and behaves like a normal AI chat response. You can review the gathered information, read the analysis, export the output, and ask follow-up questions. The main use case is long-running deep research, such as launching several hour-long tasks before stepping away and returning when the completed research is ready.