What is Self-Learning?
Self-Learning is how your Agent Stack gets better over time without you rewriting prompts by hand. When something the AI said wasn’t quite right, anyone on your team flags the response and adds a note. Atender groups similar flags together, generates concrete proposed fixes, and lets you approve and publish those fixes from one place.
The result: a feedback loop that turns “the AI keeps getting this wrong” into a system-prompt edit, a routing tweak, or a new knowledge article — applied to your stack the moment you click Approve.
How the loop works
- Flag — anyone reviewing a conversation flags an AI response that missed the mark. They pick a feedback mode (fix this specific message or learn from this in general) and tag what was wrong: missing info, too verbose, incorrect, wrong tone, routing error, knowledge gap, or other.
- Group — Atender clusters semantically similar flags into Feedback Groups. If five different customers all asked about returns and the AI gave a fuzzy answer five times, those five flags collapse into one group with a clear summary.
- Propose — for each group, Atender generates one or more proposals — concrete changes you could apply. Possible proposal types are summarized below.
- Approve — you review each proposal in Staged Changes, approve or reject. Approved proposals are deployed to your stack and the change is recorded in the Audit Log. If a deployed change makes things worse, revert it from the same audit log.
What can be proposed
- Prompt edit — Updates a specialist agent’s system prompt to handle the recurring case better.
- KB article — Suggests a new or improved knowledge base article that fills a recurring knowledge gap.
- Handbook article — Suggests an internal-handbook entry to guide the AI’s behavior on the topic.
- Routing update — Adjusts the router so similar future questions reach the right specialist.
- Temperature adjustment — Tweaks the AI’s creativity / variance for the specialist when its style is off.
Each proposal includes a diff summary showing exactly what would change before you approve it — no black-box updates.
Three tabs in Settings → Self-Learning
- Feedback Groups — Browse the clustered feedback. Each group shows what’s been flagged and how often. Drill in to see individual annotations.
- Staged Changes — The queue of proposals awaiting approval. This is where most of your day-to-day Self-Learning work happens.
- Audit Log — Every approved, deployed, or reverted change. Shows who reviewed it, when, and what changed.
The Staged Changes tab is only visible to users with full access to Self-Learning; review-only users see Feedback Groups and Audit Log.
How it relates to manual flagging
Self-Learning shares the same flagging mechanism as the test sandbox and live conversations. A flag from a real customer conversation, a flag from your test session, and a flag from a Monitor review all flow into the same Feedback → Group → Proposal pipeline. The more flags you collect, the better the groups and proposals get.
What it’s not
- Not automatic — proposals don’t auto-deploy. Every change goes through your approval.
- Not cross-tenant — the AI never learns from other tenants’ data. Improvements made in your stack apply only to your stack.
- Not retroactive — past conversations aren’t replayed against new prompts. Changes apply going forward.
Where to start
- See it in action: open Settings → Self-Learning and browse the Feedback Groups tab — even on a fresh stack, the empty state explains the workflow.
- Flag your first response: Flag and tune bad responses
- Approve a proposal: Review and approve a proposal