Recruiting automation

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 recruiting operating system that captures candidates from WhatsApp, referral, or inbound application; screens them against role criteria using an AI agent; creates a structured review task for the recruiter; and logs every decision in Max Activity — turning a scattered hiring process into a visible, auditable pipeline.

After deployment: a candidate enters the system, Frontline creates a Person record and a candidate Table row with role criteria comparison, the recruiter receives a structured summary and review task, and the decision state is tracked from first contact through offer.

When to use this

Your hiring process is managed in spreadsheets and email threads with no visibility into where each candidate stands.

Recruiters spend time on candidates who clearly don't fit the role before the screening call.

Hiring managers complain that candidates reach them without context.

You need an auditable trail of screening decisions for compliance or equity review.

System components

CRM Person record: candidate name, phone, role applied for, source, screening status, recruiter owner.

Studio Table: role criteria (required skills, experience range, location, role-specific criteria).

Studio workflow: triggered on new WhatsApp application message or form submission.

AI screening agent: compares candidate profile against role criteria, returns fit summary with signals.

Conditional Routing: high-fit → recruiter review task; low-fit → polite rejection; missing info → request more.

Max Activity: screening summary, fit signals, recruiter decision, next step.

Step-by-step implementation

1. Create a Studio Table called 'Role Criteria'. Add columns: role_name, required_skills, experience_min, experience_max, location_requirement, disqualifying_criteria.

2. Add rows for each open role. Fill in the criteria precisely — the AI will use this to screen.

3. Create a CRM Person record template for candidates. Add fields: role_applied, source, screening_status (New/Screened/Recruiter Review/Offer/Rejected), recruiter_owner.

4. Create a Studio workflow. Trigger: new WhatsApp message on the recruiting phone number.

5. Add an AI capture node to extract: name, role they're applying for, brief background summary from their message.

6. Add a Table lookup node to fetch the role criteria for the role they mentioned.

7. Add AI screening agent node. Pass: candidate message, their background, the role criteria. Return: fit assessment with specific signals.

8. Add Conditional Routing based on fit: High (≥ 70% criteria met) → create recruiter review task; Low → polite rejection message; Missing info → ask for more details.

9. High-fit branch: create Person record, set screening_status = 'Recruiter Review', create Max Task for recruiter with AI screening summary, write Max Activity.

10. Low-fit branch: send polite message ('We'll keep your profile for future roles'), write Max Activity with rejection reason (internal, not sent to candidate).

11. Test with 3 different candidate profiles: one strong fit, one clear mismatch, one incomplete profile.

Agent prompt

You are a recruiting screening agent. Compare this candidate's profile against the role criteria and produce a structured fit assessment.

You will receive: the candidate's message or resume summary, and the role criteria table (required skills, experience range, location, disqualifying criteria).

Produce a fit assessment: list which criteria they meet, which they don't, and which are unknown. Include a fit score (0-100) and a one-paragraph summary for the recruiter.

Be specific. Do not say 'candidate has relevant experience' — say 'candidate mentions 3 years in B2B SaaS sales which meets the 2-year minimum requirement'.

Output JSON: { fit_score: number, criteria_met: string[], criteria_not_met: string[], unknown_criteria: string[], summary: string, recruiter_note: string }.

Workflow logic

Fit score ≥ 70: create Person record, set status = Recruiter Review, send recruiter task with summary, write Max Activity.

Fit score 40–69: create Person record as 'Screened - Borderline', write Max Activity, add to recruiter review queue with lower priority.

Fit score < 40: send polite rejection WhatsApp message, write Max Activity with internal rejection reason (not the score), do not create a recruiter task.

Missing criteria (incomplete profile): ask candidate for the missing information (one question). Re-screen after reply.

Disqualifying criterion present: automatically move to rejected regardless of overall score. Disqualifying criteria are non-negotiable.

Final operating state

Every candidate who contacts via WhatsApp is captured in CRM with role applied, source, and initial screening status.

Recruiters receive Max Tasks only for candidates who meet the minimum criteria — no manual pre-filtering required.

Every screening decision is logged in Max Activity with the fit signals used, so hiring managers can audit the process.

Borderline candidates are held in a review queue rather than auto-rejected, preserving human judgment for edge cases.

Troubleshooting

AI screening too lenient (too many high-fit scores): add more specific required criteria to the role criteria table. The AI matches what you define.

Candidates rejected for unclear reasons: check that the reject branch writes the internal reason to Max Activity with the specific unmet criteria.

Same candidate applying multiple times: add duplicate detection — check for existing Person by phone before creating a new record.

Role criteria table not found: verify the table lookup uses the exact role name from the candidate's message. Add fuzzy matching or a role name normalization step.

Operational playbook

Use Recruiting automation 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