Automate AI-Powered CV Screening with n8n & Google Gemini

1. Opening Problem Statement

Meet Sarah, an HR manager at a fast-growing software development company. Each week, she receives hundreds of resumes for the open Software Engineer position. Sarah spends hours manually reviewing each CV, trying to determine if candidates match the job requirements. On top of that, she needs to send confirmation emails to candidates and notify her HR team. This tedious process drains Sarah’s time, causes delays in hiring decisions, and risks missing top talent due to human error or fatigue.

Imagine Sarah losing 10+ hours a week just sorting through CVs and sending emails manually. This delay in screening candidates slows down the hiring pipeline, potentially costing the company valuable opportunities to onboard skilled engineers faster.

2. What This Automation Does

This automation workflow using n8n with the Google Gemini (PaLM) AI model transforms Sarah’s hiring process by:

• Automatically receiving CV submissions via a custom application form.
• Extracting the uploaded resume from PDF format into JSON for AI analysis.
• Using advanced AI to evaluate candidate CVs against the Software Engineer job description, generating a compatibility rating and interview recommendation.
• Logging detailed candidate information and AI ratings into a Google Sheet for easy tracking and future review.
• Sending instant, personalized confirmation emails to candidates acknowledging receipt of their CVs.
• Alerting the HR team with a detailed email including the AI analysis and candidate details to speed up decisions.

By automating these tasks, Sarah saves over 10 hours weekly, eliminates errors in manual data processing, and gains data-driven clarity on candidate fit.

3. Prerequisites ⚙️

• n8n account – to create and run the workflow.
• Google Gemini (PaLM) API credentials – to access the AI model for CV analysis.
• Gmail account – configured for sending confirmation and HR notification emails.
• Google Sheets account – to store candidate data and AI ratings.
• Ability to receive file uploads in n8n via Form Trigger node.

Optionally, if you want to host n8n yourself for full control and privacy, explore self-hosting options like Hostinger.

4. Step-by-Step Guide to Build AI CV Screening Workflow

Step 1: Set up the Application Form trigger

Navigate to Triggers → Form Trigger in n8n.

Create a form titled Application for Software Engineer Position with fields: Full Name (required), E-mail (required), Expectation (salary range, required), Linkedin Profile (required), and a file upload for Your Resume/CV (accepts only PDF, required).

This node waits for candidates to submit applications via the form webhook.

Common mistake: Forgetting to set the file upload field to accept only PDFs, which may cause parsing errors later.

Step 2: Extract PDF resume content as JSON

Add an Extract From File node connected to the form trigger.

Configure it for the operation PDF and specify the binary property as Your_Resume_CV to extract the uploaded PDF into JSON-readable text.

This prepares the resume content for AI processing.

Common mistake: Not matching the binary property name exactly, which leads to an empty output.

Step 3: Analyze resume using Google Gemini Chat Model

Insert the Google Gemini Chat Model node.

Authenticate with your Google Palm API credentials.

This node receives the extracted resume text and job description prompt, then runs a detailed compatibility analysis, returning a structured rating and recommendation.

Key parameter: Provide the customized prompt that instructs the AI to analyze the candidate’s fit for a Software Engineer role within 75 words.

Common mistake: Omitting the job description or using an incompatible prompt format.

Step 4: Using AI Analysis & Rating Node for final processing

Connect the output from Google Gemini Chat Model to the Chain LLM node labeled Using AI Analysis & Rating.

This node structures the AI’s text output, ensuring it’s clean and ready for storage.

Common mistake: Not passing the AI output text properly, causing null or incomplete data downstream.

Step 5: Append candidate info and AI rating to Google Sheets

Use the Google Sheets node connected to the AI Analysis output.

Configure it to append data to your Google Sheet document under a sheet named পত্রক1.

Map candidate fields: CV filename, E-mail, Linkedin, AI Rating, Full Name, Expectation.

Common mistake: Incorrect mapping or missing sheet access permissions can cause failures.

Step 6: Inform HR of new CV received

Add a Gmail node titled Inform HR New CV Received.

Set recipient to your HR email address.

Compose the message including candidate details and AI review results to help HR make faster decisions.

Common mistake: Not using dynamic expressions for candidate info, resulting in generic emails.

Step 7: Send confirmation email to candidate

Add a Gmail node Confirmation of CV Submission connected after the HR notification.

Configure recipient as the applicant’s email.

Write a polite message thanking them for submission, reinforcing positive candidate engagement.

Common mistake: Forgetting to map the candidate’s email or name dynamically in email fields.

5. Customizations ✏️ Specific to This Workflow

• Change the job title for AI analysis: Modify the prompt in the Using AI Analysis & Rating node by replacing “Software Engineer” with any other role you are recruiting for.
• Add additional candidate fields: Extend the form node by including fields like phone number or portfolio URL, then map them through subsequent nodes.
• Use another AI model: Swap out the Google Gemini Chat Model node with a different AI service integrated with n8n, updating credential and prompt details accordingly.
• Customize confirmation email text: Tailor the message content and subject line in the Confirmation of CV Submission Gmail node to align with your company’s tone.
• Automate LinkedIn profile scraping: Add a separate HTTP request node to fetch public LinkedIn data, enriching candidate profiles automatically.

6. Troubleshooting 🔧

Problem: “No binary data found for extraction” after resume upload.
Cause: Incorrect field name or file type uploaded.
Solution: Verify the form field is named exactly “Your Resume/CV” and configured to accept PDFs only. Ensure the file uploads correctly before extraction.

Problem: “Google API authentication failed” in Gemini Chat node.
Cause: Invalid or expired API credentials.
Solution: Reauthenticate your Google Gemini API credentials in n8n and ensure billing is active on your Google Cloud account.

Problem: Emails sent without dynamic candidate data.
Cause: Incorrect expression syntax or missing input data.
Solution: Double-check Gmail node’s “Send To” and “Message” fields use correct expressions like {{ $('Application Form').item.json['E-mail'] }}.

7. Pre-Production Checklist ✅

• Test the form by submitting sample CVs with all required fields filled and a PDF resume attached.
• Verify the PDF extraction outputs readable JSON text for the AI analysis.
• Confirm Google Gemini node generates a valid compatibility rating and recommendation.
• Ensure Google Sheets receives appended candidate data correctly.
• Check both HR notification and candidate confirmation emails arrive with accurate customized content.
• Backup your Google Sheets document before starting live submissions.

8. Deployment Guide

Activate the workflow in n8n by toggling the active switch.

Monitor the execution logs for errors during early runs. Adjust node configurations if needed.

Regularly back up your Google Sheets to prevent data loss.

Optionally, add error handling nodes or notifications to catch submission failures.

9. FAQs

Q: Can I use another AI model like OpenAI instead of Google Gemini?
A: Yes, you can replace the Google Gemini node with other AI nodes, but you need to adjust prompt formats and credentials accordingly.

Q: Does this workflow consume my Google API quota quickly?
A: Frequent AI calls can add up; monitor your usage in Google Cloud and optimize by batching if needed.

Q: Is candidate data securely handled?
A: Data is processed within your n8n instance and Google services under your control. Always follow best security practices when managing credentials.

10. Conclusion

By creating this n8n AI CV Screening workflow integrated with Google Gemini, you have automated the labor-intensive resume screening process that Sarah once struggled with. You’ve enabled rapid, AI-driven candidate evaluation, timely communications, and organized data logging — saving 10+ hours weekly and improving hiring quality.

Next, consider automating interview scheduling, adding candidate feedback collection, or integrating with your ATS software to continue enhancing recruitment efficiency.

With this powerful, beginner-friendly guide, you’re well on your way to mastering AI-powered hiring automation through n8n!