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

# Candidate Evaluation

> Score candidates against a job description from CVs, assessments, and interview notes, with a ranked executive summary

**Goal:** turn a stack of CVs, assessment results, and interview transcripts into a consistent, ranked evaluation of every candidate against the actual requirements of the role.

<Frame>
  <iframe style={{ width: '100%', aspectRatio: '16/9' }} src="https://www.youtube.com/embed/gdFQy8-2pFw" title="How to automate candidate evaluation with VirtualBrain" frameBorder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowFullScreen />
</Frame>

Recruiting decisions usually rest on whoever interviewed last and remembered most. The Candidate Evaluation App reads everything, every transcript and every test score, and evaluates each candidate against the same competency framework, derived from the job description itself.

## Setup

Upload the full evaluation file for each candidate:

* The **job description** for the role
* Each candidate's **CV**
* **Technical assessments** or assessment center results
* **Interview transcripts or interview notes**

In the video, five candidates are evaluated at once; the App handles a substantial amount of test data per candidate.

## How it works

<Steps>
  <Step title="Launch the App on your candidate documents">
    Upload all candidate materials and the job description, then start the App.
  </Step>

  <Step title="Competency extraction">
    The first block examines the job description, derives which competencies to evaluate (in the demo, customer obsession and strategic relationship management), and defines the required mastery level for each.
  </Step>

  <Step title="Document review per competency">
    For each relevant competency, VirtualBrain reviews the available documents, including the technical documents and the job description.
  </Step>

  <Step title="Comprehensive analysis">
    The App reviews each candidate's technical test scores and interview performance in detail, across all five profiles.
  </Step>

  <Step title="Scoring">
    Each candidate is assessed individually on every dimension identified during the evaluation setup, with an overall score plus key strengths and watch-outs per profile.
  </Step>

  <Step title="Executive summary">
    The final block ranks all candidates by overall performance, emphasizes key interview insights, and identifies development priorities for each selected profile.
  </Step>
</Steps>

## What you get

A 360-degree view of each candidate's performance and potential development areas: per-dimension scores, strengths and watch-outs per profile, and a ranked executive summary. It serves as a foundation for the hiring decision, and as the starting point for the selected candidate's onboarding development plan.

## Pro tips

* **The job description drives everything.** The competency framework is extrapolated from it, so a precise job description produces a precise evaluation.
* **Feed it complete transcripts, not summaries.** The analysis is only as thorough as the interview record; raw transcripts preserve the evidence behind each score.
* **Treat the ranking as decision support, not the decision.** Review the strengths and watch-outs behind each score with the hiring panel before committing.
* **Reuse the development priorities.** They translate directly into the first draft of the new hire's onboarding plan.

## Related

<CardGroup cols={2}>
  <Card title="Using App Templates" icon="grid-2" href="/features/apps/using-templates">
    Find and launch the Candidate Evaluation template
  </Card>

  <Card title="Building Apps" icon="wrench" href="/features/apps/building-apps">
    Adapt the scoring blocks to your firm's competency model
  </Card>

  <Card title="Organizing Documents" icon="folder-tree" href="/features/brains/organizing-documents">
    Keep candidate files structured for clean evaluations
  </Card>

  <Card title="Sourced Answers" icon="quote-left" href="/features/chats/sourced-answers">
    Trace every score back to the underlying transcript or test
  </Card>
</CardGroup>
