Candidate screening

Concepts covered

Step breakdown

1. Define the system to deployStart from this HR operating system and confirm the business outcome, source signals, owners, and review gates.
2. Create the Frontline assetsBuild the workflow canvas, agent prompt, channels, CRM records or tables, routing logic, and Max Activity evidence described in the blueprint.
3. Launch the narrow versionDeploy the smallest reliable version first, keep human review visible, then use activity and analytics to expand the system.

What you will build

A candidate screening system that uses AI to compare candidate profiles against structured role criteria, produces an explainable fit summary for recruiters, enforces review gates for borderline and sensitive cases, and logs every screening decision in Max Activity for audit.

After deployment: every candidate entering the pipeline is screened against the role criteria before reaching a recruiter. Recruiters receive a structured summary — not a raw resume — and make decisions from a consistent information base.

When to use this

Recruiters are reviewing many unqualified candidates before finding a fit.

Screening decisions are inconsistent across recruiters or across time for the same role.

You need to document screening decisions for equity, compliance, or legal review.

Hiring velocity is slow because the first-pass review is manual and time-consuming.

System components

Studio Table: role criteria with required skills, experience range, location, and disqualifiers.

Playbook: candidate screening process steps — what to extract, what to assess, how to weight criteria.

AI screening agent: reads candidate profile and role criteria, produces structured fit summary.

CRM Person record: candidate fields, role applied, screening status, fit score, recruiter owner.

Review gate: borderline candidates require recruiter decision before advancing.

Max Activity: screening summary, criteria comparison, fit score, recruiter decision, next step.

Step-by-step implementation

Agent prompt

You are a candidate screening agent. Evaluate this candidate against the role criteria and produce a structured fit assessment for the recruiter.

Use the screening playbook attached to determine what to extract and how to weight each criterion.

Extract from the candidate profile: years of relevant experience, specific skills mentioned, location, role-specific requirements.

Compare each extracted value against the role criteria. Be specific and evidence-based. Quote what the candidate said or listed.

Output JSON: { fit_score: 0-100, criteria_met: [{ criterion: string, evidence: string }], criteria_not_met: [{ criterion: string, reason: string }], unknown_criteria: string[], disqualifiers_found: string[], recruiter_summary: string }.

Workflow logic

Fit score ≥ 70, no disqualifiers: advance to recruiter review. Create Max Task with structured summary.

Fit score 40–69: borderline review gate. Recruiter must make an explicit decision (advance or reject) before workflow continues.

Fit score < 40 OR disqualifier found: disqualify. Update Person status. Write Max Activity. Human-reviewed rejection message sent later.

Unknown criteria (information not in profile): create Max Task asking recruiter to verify the specific unknown criteria before advancing.

Recruiter advances a borderline candidate: log the advancement reason in Max Activity. This creates the audit trail.

Final operating state

Every candidate is screened against the same role criteria using the same process.

Recruiters receive Max Tasks only for candidates who meet the minimum threshold — clear mismatches are filtered before human review.

Borderline candidates are not auto-rejected — they go to a human review gate.

Max Activity contains a complete audit trail: screening criteria, fit score, evidence for each criterion, and recruiter decision.

Troubleshooting

Fit score too high for clearly unqualified candidates: verify the role criteria table has specific, measurable requirements. Vague criteria produce inflated scores.

Borderline queue growing too large: recalibrate the threshold. If 40% of candidates are borderline, the criteria may be too broad or the scoring too permissive.

Recruiter not receiving Max Tasks: confirm the task creation node uses the correct recruiter_owner field from the Person record.

Disqualified candidates receiving premature rejections: verify the disqualify branch does NOT trigger a Send Message node. Rejections must be sent manually after human review.

Operational playbook

Use Candidate screening as part of the Frontline Solutions HR operating loop: inspect the current product state, confirm the source context, and decide what should happen next.

The goal is not to memorize screens. The goal is to understand how the product surface supports repeatable work, AI assistance, and accountable handoff.

Best practices

Start with the operational job before changing configuration. Name the owner, define the trigger or source context, and decide how the result should be reviewed.

Prefer narrow, inspectable setups over broad automation. Teammates should be able to explain why the system took an action from the visible product state.

Platform layers involved

Studio defines the workflow and AI agent behavior. Channels capture the customer interaction. CRM provides customer memory. Max Activity shows what the system did and what needs follow-up.

Use the solution page as the business-facing map, then open the related product tutorials when you need configuration detail.

Outcome metrics

Track a small set of operational signals: response time, handoff rate, completion rate, escalation quality, CRM field completeness, reply rate, and repeated failure patterns.

The metric should reflect the business outcome, not only whether the automation ran.

FAQs

Related workflows