Start here as a trainer

You are a Trainer. Responsibility for the learning process is yours. Field experts decide whether a scenario is professionally accurate; you decide whether it is right for this cohort right now, and you render the canonical verdict on every piece of competency evidence the AI extracts. Nothing reaches a learner record without your decision.

Your home shows

When you sign in, your home opens on the Bandeja and surfaces the work that needs you:

• The Bandeja — the queue of pending AI signals waiting on your verdict, oldest first. After 7 days unresolved, your org admin sees an escalation.
• Engagements in flight — current learner engagements you can drop into mid-dialogue, or read after the fact.
• Assignments past due — scenarios you assigned that the learner has not started by the due date.
• Recent disagreements — verdicts where you diverged from the AI signal. Pattern-spotting, not punishment.

How the work flows

Each closed engagement produces one or more AI signals — one per sub-competency the dialogue exercised. Each lands in the Bandeja. You open it, read the signal and the cited dialogue snippet, and choose Accept, Adjust, or Reject. Your verdict moves the learner’s radar; the AI signal alone never does.

See evidence and verdicts for the data shape and the eight sub-competencies for what each dimension measures.

Your verdict is canonical

The AI is a sparring partner that proposes; you are the judge that decides. Pondara guarantees, at the database layer, that no actor other than a trainer’s stamped verdict can move a learner’s competency vector.

Where to next

Triage the bandeja Work through pending evidence in priority order.

Render a verdict Accept, Adjust, or Reject — with the undo window.

Assign scenarios Push a validated scenario to one or more learners.

Learner detail page Read a learner's radar, history, and pending verdicts.

You can also disagree with the AI, review a dialogue, or manage overdue work.