Skills, Experience, and Education assessments (AI-driven)

CVViZ's AI breaks candidate evaluation into Skills, Experience, and Education scores — what each shows and how to read them.

What these assessments are

Skills, Experience, and Education assessments in CVViZ are AI-driven evaluations of a candidate's profile against the job. They're not manual rubrics interviewers fill out — that's the Interview Scorecard. The assessments are computed automatically when AI evaluation runs on a candidate.

Each is a separate component on the candidate profile, plus an Overall Assessment that aggregates all three.

Where to see them

Open a candidate. The evaluation components surface on the candidate's profile / evaluation view (component path: src/components/CandidateEvaluation/). Each block shows a score and the underlying matching detail.

Skills Assessment

What it shows:

  • Skills score — a percentage (0–100) for how well the candidate's parsed skills match the job's required and preferred skills.
  • Required skills match — which mandatory skills the candidate has.
  • Preferred skills match — which nice-to-haves the candidate has.
  • Missing required skills — mandatory skills the candidate is missing.

This is a structured view of skill fit beyond the simple letter grade.

Experience Assessment

What it shows:

  • Experience score — percentage match.
  • Total years experience — parsed from the CV.
  • Relevant years experience — years that look relevant to this specific role.
  • Role alignment — how closely past roles map to the open one.

The "relevant years" is more useful than total years — a 15-year veteran with no recent backend experience might score lower than an 8-year specialist.

Education Assessment

What it shows:

  • Education score — percentage match.
  • Education level match — does the candidate's education meet the job's level requirement?
  • Field of study alignment — does the major align?

Overall Assessment

The OverallAssessment component aggregates the three above and surfaces a combined view. This is the AI's holistic read on candidate fit.

How the AI scores work

These assessments are part of CVViZ's AI evaluation pipeline (turned on workspace-wide via Settings → Automations). The signals the AI uses:

  • The job's description, mandatory skills, qualifications, and experience range.
  • The candidate's parsed CV (skills, work history, education).
  • Any Benchmark Resumes on the job — these are the strongest accuracy signal you can give the AI. See Benchmark Data: training AI matching with sample resumes.

Manual evaluation vs AI assessment

To avoid confusion:

  • Skills / Experience / Education Assessments = AI-driven, automatic, computed against parsed CV data.
  • Interview Scorecard = manual, filled by interviewers after meeting the candidate. See Interview scorecards: building and assigning.

Use both: the AI assessments help you triage; the Interview Scorecard captures human judgment after the conversation.

Tips for accuracy

  • Mark mandatory skills correctly. The Skills Assessment leans heavily on which skills you marked as mandatory.
  • Write a thorough JD. The AI uses the job description as a primary signal.
  • Add Benchmark Resumes to the job — the strongest input for accurate matching.
  • Sample manually. Spot-check 10–20 candidates' assessments to confirm the bar matches yours; if it doesn't, the fix is usually in the JD or benchmarks.