HR & Recruiting

Screen Resumes with Vision AI

Turn a resume PDF into an image and let a vision model assess fit, which resists hidden text meant to trick AI screeners.

The Problem

Some applicants hide invisible instructions in resumes to manipulate AI screening. Text-based parsing reads those instructions as real content.

How This Template Solves It

A recruiter pulls a candidate's resume from Google Drive or an applicant tracking system. The PDF is converted to an image, and Gemini reads what a person would see and judges fit for the role, so hidden text has no effect.

How It Works

  1. 1Downloads the resume PDF from Google Drive.
  2. 2Converts the PDF to an image with Stirling PDF and reduces the resolution.
  3. 3A multimodal Gemini model evaluates the image against the role.
  4. 4Returns the assessment for the next step in your hiring process.

What You Need

  • Google Drive or an applicant tracking system
  • A Gemini API key
  • Stirling PDF

What Stays Under Your Control

  • A recruiter decides whether to move the candidate forward.

Watch Out For

  • Reading resumes as images does not remove bias. Keep human review.
  • Candidate data is sensitive. Self-host the PDF converter where you can.

Adapted from a community workflow template. Before publishing we removed credentials, account and file IDs, email addresses and sample data. The original author's setup notes are kept inside the workflow file.

More HR & Recruiting Templates

HR & Recruiting

Job Applicant Scoring and Interview Scheduling

Score job applicants against the job description with AI, shortlist strong ones, generate interview questions and book a call.

Starts on: Form submission

AirtableEmail (SMTP)Google Calendar+2 more