Liya Engine

Candidate Prescreen

Automatically prescreen a candidate against a job description — scoring fit, surfacing strengths and concerns, and recommending whether to proceed.

POST /v1/hiring/candidate-prescreen

Employer-facing intent. Evaluates a candidate's profile against a job description and returns a structured prescreening report: overall scores, a proceed/decline recommendation, cited strengths and concerns, risk flags, and recommended interview focus areas.


Required inputs

FieldTypeNotes
input.user_idstringYour platform's user ID (e.g. recruiter ID)
input.profileProfileInputCandidate's structured profile
input.job_descriptionJobDescriptionInputRole requirements

Optional inputs

FieldTypeNotes
input.resumeResumeInputResume text to supplement profile data

Example request

curl -X POST https://api.liyaengine.ai/v1/hiring/candidate-prescreen \
  -H "x-api-key: $LIYA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "user_id": "recruiter_456",
      "profile": {
        "name": "Alex Johnson",
        "years_of_experience": 9,
        "experience_level": "senior",
        "skills": ["Python", "machine learning", "PyTorch", "AWS", "SQL"],
        "work_experiences": [ ... ]
      },
      "job_description": {
        "title": "Senior ML Engineer",
        "description": "Join our ML platform team to build and deploy production models...",
        "required_skills": ["Python", "PyTorch", "AWS", "MLOps"],
        "required_experience_years": 5
      }
    }
  }'

Response

{
  "success": true,
  "data": {
    "output": {
      "overall_score": 88,
      "fit_score": 85,
      "skill_score": 92,
      "experience_score": 90,
      "confidence": { "score": 0.93, "level": "high" },
      "recommendation": "strong_proceed",
      "recommendation_rationale": "Candidate exceeds core technical requirements with 9 years of relevant experience and direct PyTorch and AWS skills. Minor gap in MLOps tooling is addressable.",
      "strengths": [
        {
          "strength": "Deep PyTorch expertise",
          "evidence": "Led model training infrastructure at previous employer",
          "relevance_to_role": "Directly maps to core technical requirement"
        }
      ],
      "concerns": [
        {
          "concern": "No explicit MLOps tooling mentioned",
          "severity": "minor",
          "evidence": "Resume lists model training but not deployment pipelines",
          "suggested_probe": "Walk me through how you've handled model deployment and monitoring in production."
        }
      ],
      "risk_flags": [],
      "interview_recommendations": {
        "recommended_interview_type": "Technical + system design",
        "focus_areas": ["MLOps pipeline ownership", "Scale of previous deployments"],
        "questions_to_ask": [
          "Describe your largest production ML system end-to-end.",
          "How have you handled model drift in production?"
        ]
      },
      "executive_summary": "Strong candidate with 9 years of directly relevant experience. Skill match is excellent. Recommend fast-tracking to technical interview."
    }
  }
}

Recommendation values

ValueMeaning
strong_proceedHighly qualified — prioritise in pipeline
proceedMeets requirements — advance to next stage
reviewPartial match — human review recommended
declineDoes not meet minimum requirements

Concern severities

ValueMeaning
blockerDisqualifying if confirmed
significantMeaningful gap to probe in interview
minorSmall gap unlikely to affect performance

On this page