AI UX PlaygroundNewsletterJoin 2K+ AI designers and PMs on Substack. New teardowns, patterns, and prompts as they drop.

Agents and workflows · Updated July 6, 2026

Thread / Session

A thread (or session) is a bounded conversation or task run with its own history, memory scope, and artifacts, distinct from other parallel chats.

Users need to know what persists inside a thread vs across the product, especially when memory, connectors, and context windows apply per session.

What it means

The product groups messages, tool calls, and outputs under one identifier; closing, renaming, or forking a thread changes what context future turns include.

Why designers should care

Thread UX needs titles, last-active timestamps, memory scope indicators, and clear “new chat” semantics so users do not accidentally mix unrelated work.

Example

A research copilot lists threads by project name; each shows token usage, pinned sources, and “Memory: this thread only” so users know prior chats are not leaking in.

Common mistakes

  • Infinite single thread with no way to start fresh without losing history.
  • Unclear whether connectors or memory apply globally or per thread.
  • No thread titles, so users cannot find last week’s PRD draft.

Frequently asked questions

What is Thread / Session?

A thread (or session) is a bounded conversation or task run with its own history, memory scope, and artifacts, distinct from other parallel chats.

Why should designers care about Thread / Session?

Thread UX needs titles, last-active timestamps, memory scope indicators, and clear “new chat” semantics so users do not accidentally mix unrelated work.

What are common mistakes with Thread / Session?

Infinite single thread with no way to start fresh without losing history; Unclear whether connectors or memory apply globally or per thread; No thread titles, so users cannot find last week’s PRD draft.