AutomationsHiringHRAI Tools

The $5 AI Hiring Assistant: Auto-Screen Applicants Using Make + OpenAI

5 min read

Save hours reading resumes. Build an automated system that screens applicants, scores them against your criteria, and summarizes the best candidates.

Hiring is the most important thing you do, but the process is broken.

You post a job. You get 200 applications.

* 100 are spam.

* 50 are unqualified.

* 40 are "maybe".

* 10 are great.

Finding those 10 requires reading 200 resumes. It takes days.

What if you had an AI recruiter who read every single application instantly, compared it to your job description, and gave you a 1-sentence summary and a "Fit Score"?

You would only have to look at the top 10%.

In this guide, we will build an AI Hiring Assistant that lives in Airtable and runs on Make.com.

Time to build: 40 minutes Tools required: Airtable, Make.com, OpenAI API, Tally Forms (or Typeform/Google Forms)

Why Automate Screening?

  • Speed: Great candidates are off the market in days. If you take a week to screen, you lose them. This system screens them in seconds.
  • Consistency: AI judges the first applicant and the last applicant with the exact same criteria.
  • Sanity: You stop drowning in PDFs and start focusing on interviews.
  • The Workflow Blueprint

  • Input: Candidate fills out a form (Tally/Typeform).
  • Trigger: Make.com detects the new submission.
  • Analyze: OpenAI compares the candidate's answers/resume text against your Job Description.
  • Score: AI assigns a score (1-10) and writes a "Why?" summary.
  • Store: Data is saved to Airtable, sorted by score.
  • 1

    Application

    Form Submitted

    2

    Make.com

    Extract Data

    3

    AI Analysis

    Score & Summarize

    4

    Airtable

    Rank Candidates

    Step-by-Step Setup Guide

    Phase 1: The Application Form

    Use Tally.so (it's free and beautiful).

    Create fields for:

    * Name

    * Email

    * LinkedIn URL

    * "Why are you a good fit?" (Long text)

    "Paste your Resume/CV text here" (Long text - Pro tip: Parsing text is easier than parsing PDF files for beginners*)

    Phase 2: The Database (Airtable)

    Create a table "Candidates" with columns:

    * Name

    * Email

    * Application Text (Long Text)

    * AI Score (Number 1-10)

    * AI Summary (Long Text)

    * Status (Single Select: To Review, Interview, Reject)

    Phase 3: The Automation

  • Trigger: Make.com -> Tally "Watch New Responses".
  • Action: OpenAI -> "Create a Completion".
  • * System Prompt:

    > You are an expert HR recruiter. I will give you a candidate's application and a job description.

    > 1. Analyze how well they match.

    > 2. Assign a score from 1-10 (10 is perfect match).

    > 3. Write a 2-sentence summary of their strengths/weaknesses.

    > Output format: JSON {"score": 8, "summary": "Strong experience in React, but lacks the requested Python knowledge."}

    * User Prompt:

    > Job Description: [Paste your JD here]

    > Candidate Application: {{Why_Fit}} {{Resume_Text}}

  • Action: Airtable -> "Create a Record".
  • * Map Name, Email from Tally.

    * Map Score and Summary from OpenAI output.

    Phase 4: The Review

    Now, when you open Airtable, you don't see a blank list. You see a sorted list.

    * Candidate A: Score 9/10. "Perfect match, 5 years exp."

    * Candidate B: Score 2/10. "Applied for wrong role."

    You start calling Candidate A immediately.

    Advanced Enhancements

    1. Auto-Email for Top Candidates

    Add a filter: If Score >= 9, send an automatic email:

    "Hi [Name], your application stood out. Here is a link to book a 15-min intro call."

    This is how you win talent—speed.

    2. Slack Notifications

    Send a message to #hiring only for candidates with Score > 7.

    "New strong candidate: [Name] (Score: 8/10). [Link to Airtable]"

    3. Resume PDF Parsing

    If you want to accept PDF uploads, add a module like PDF.co "Parse Document" before the OpenAI step to extract the text layer from the PDF file.

    I recently hired a video editor. I asked for a link to their portfolio and a description of their editing style.

    * AI Prompt: "Check if they mentioned 'Premiere Pro' and if their style description matches 'fast-paced, retention editing'."

    * Result: I filtered 50 applicants down to 6 interviews in 10 minutes.

    Conclusion

    Hiring is about finding the signal in the noise. AI is the ultimate noise-canceling headphone. It doesn't replace your judgment—it just ensures you spend your time judging the right people.

    Ready to build this?

    * Get your free Make.com account here

    * Try Tally Forms


    Disclaimer: This article contains affiliate links. If you purchase through these links, we may earn a commission at no extra cost to you.

    Frequently Asked Questions

    Is this legal/ethical?

    AI should be used as a 'first pass' filter, not the final decision maker. Always manually review highly-scored candidates. Ensure your prompt instructs the AI to ignore race, gender, and age to reduce bias.

    Can it read PDF resumes?

    Yes. Make.com has modules (like 'PDF.co' or 'Google Cloud Vision') to extract text from PDFs. You can then feed that text into OpenAI.

    How accurate is the scoring?

    Surprisingly good if your job description is clear. It's consistent. It won't have a 'bad day' or get tired after reading 50 resumes.

    Can I auto-reject low scores?

    Technically yes, but it's risky. A better approach is to auto-email them saying 'We have received your application' and manually trigger the rejection later.

    Share this article

    Want more tips like this?

    Explore our blog for more guides on AI automation for solopreneurs.

    Browse All Articles