> ## Documentation Index
> Fetch the complete documentation index at: https://docs.exterview.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# How to Run EX Avatar

> Step-by-step walkthrough for running an EX Avatar AI video interview end to end, from enabling the agent to reviewing the Avatar AI Screening Report.

A step-by-step walkthrough for running an **EX Avatar** AI video interview end to end. EX Avatar is **Available** today.

<Frame>
  <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Running EX Avatar" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
</Frame>

## Before you start

You need a configured role with EX Avatar enabled, and a candidate to invite. See [Configure](/configure/overview) and [Session Types](/deploy/session-types).

## Steps

<Steps>
  <Step title="Enable EX Avatar for the role">
    Turn on EX Avatar for the pipeline stage where the video interview should run.

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Enable EX Avatar for the role" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>

  <Step title="Confirm the questions and rubric">
    Review the questions and scoring rubric the avatar interviewer will use for this role. See [How to Set Scoring Weights](/guides/setup-scoring).

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Confirm the questions and rubric" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>

  <Step title="Invite the candidate">
    Send the invite or share the interview link, with integrity and privacy consent captured up front. See [Invite Flows](/deploy/invite-flows).

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Invite the candidate" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>

  <Step title="Candidate completes the interview">
    The candidate joins from any browser, passes the device and camera check, and completes the avatar interview.

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Candidate completes the interview" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>

  <Step title="Scoring and critic evaluation run">
    The agent scores the interview against the rubric, and the critic layer evaluates the output before it surfaces. See [AI Observability](/monitor/ai-observability).

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Scoring and critic evaluation run" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>

  <Step title="Review the Avatar AI Screening Report">
    Open the report, read the score and the evidence behind it, and check any critic flags. See [Session Reports](/monitor/session-reports).

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Review the Avatar AI Screening Report" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>

  <Step title="Make the decision">
    A human reviewer decides and advances or holds the candidate. No adverse decision is made autonomously.

    <Frame>
      <img src="https://mintcdn.com/exterviewinc/n0waW1x1D1rW5ovx/images/walkthroughs/fallback.png?fit=max&auto=format&n=n0waW1x1D1rW5ovx&q=85&s=693e5355b8f9d23fc68ed6854d5ef05f" alt="Make the decision" width="1280" height="720" data-path="images/walkthroughs/fallback.png" />
    </Frame>
  </Step>
</Steps>

## FAQs

<AccordionGroup>
  <Accordion title="What does EX Avatar produce?">
    An Avatar AI Screening Report, reviewed in Monitor.
  </Accordion>

  <Accordion title="Does it analyze the candidate's face?">
    No. Scoring is content-based, grounded in what the candidate says, not facial
    analysis or expression.
  </Accordion>

  <Accordion title="Does a person still decide?">
    Yes. The report is decision support; a human owns the outcome.
  </Accordion>
</AccordionGroup>

## Related

<CardGroup cols={2}>
  <Card title="Session Types" href="/deploy/session-types">
    The full agent catalogue with availability.
  </Card>

  <Card title="Instant Interviews (Voice / Avatar)" href="/capabilities/instant-interviews">
    The integrator view for triggering and reading back.
  </Card>

  <Card title="Session Reports" href="/monitor/session-reports">
    What the Avatar AI Screening Report contains.
  </Card>

  <Card title="EX Panel" href="/walkthroughs/ex-panel">
    The usual next step for human rounds.
  </Card>
</CardGroup>