Parse & evaluate HR candidates with GPT-4.1 and LinkedIn data in CSV/XLSX

Name: Parse & evaluate HR candidates with GPT-4.1 and LinkedIn data in CSV/XLSX
Availability: InStock
Author: Elay Guez

Parse & evaluate HR candidates with GPT-4.1 and LinkedIn data in CSV/XLSX preview

Open on n8n.io

$20/month : Unlimited workflows

2500 executions/month

Try free

THE #1 IN WEB SCRAPING

Scrape any website without limits

Try free

HOSTINGER

Early Deal
DISCOUNT 20%

Self-hosted n8n

Unlimited workflows - from $4.99/mo

Try free

#1 hub for scraping, AI & automation

6000+ actors - $5 credits/mo

Try free

Important notice

This workflow is provided as-is. Please review and test before using in production.

Overview

AI-Powered HR Candidate Evaluation Agent with LinkedIn Data Enrichment in CSV/XLSX Format

🎯 Overview

Transform your manual hiring process into an intelligent evaluation system that saves 15-20 minutes per candidate! This workflow automates the entire candidate assessment pipeline - from CSV/XLSX upload to AI-powered scoring with LinkedIn insights.

When you upload a candidate list, this workflow automatically:

📊 Converts your file into a formatted Google Sheet with RTL support
🔍 Researches each candidate's recent LinkedIn posts via Apify
🤖 Evaluates candidates using GPT-4.1 with context-aware scoring (0-100)
💬 Generates professional Hebrew explanations for each score
📈 Auto-sorts by score and applies professional formatting
⚠️ Sends error alerts to keep everything running smoothly

Cost per candidate: ~$0.05 | Time saved: 15-20 minutes each

👥 Who's it for?

HR teams drowning in candidate applications
Recruitment agencies needing consistent evaluation criteria
Hiring managers seeking data-driven candidate insights
Companies looking to scale their team
Anyone tired of manual spreadsheet juggling

⚡ How it works

Form submission triggers with CSV/XLSX upload
Google Drive stores the file and creates a new Sheet
Data extraction processes the file content
AI Agent loops through each candidate:
- Fetches up to 3 recent LinkedIn posts via Apify
- Analyzes qualifications against job requirements
- Generates evaluation score and Hebrew explanation
Sheet formatting applies filters, sorting, and styling
Error handling notifies admin of any issues

🛠️ Setup Instructions

Time to deploy: 15 minutes

Requirements:

Google account (Drive + Sheets access)
OpenAI API key (GPT-4.1 access)
Apify API key (for LinkedIn scraping)
Gmail account (for error notifications)

Step-by-step:

Import this template into your n8n instance
Configure Google credentials:
- Connect Google Drive OAuth2
- Connect Google Sheets OAuth2
Add OpenAI API key to the GPT-4.1 node
Set up Apify credentials for LinkedIn scraping
Configure Gmail for error alerts (update email in "Send a message" node)
Update folder IDs in Google Drive nodes to your folders
Test with a sample CSV containing 2-3 candidates
Activate and share the form URL with your team!

📋 Input File Format

Your CSV/XLSX should include these columns (Hebrew):

שם פרטי (First name)
שם משפחה (Last name)
חשבון לינקדאין (LinkedIn URL)
Your custom evaluation questions

🎨 Customization Options

Easy tweaks:

Scoring criteria: Modify the AI agent's system message
Language: Switch from Hebrew to any language
Scoring rubric: Adjust the 50/25/15/10 weighting
LinkedIn posts: Change from 3 posts to more/fewer
Sheet styling: Customize colors and formatting

Advanced modifications:

Add integration with your ATS (Greenhouse, Lever, etc.)
Connect to Slack for real-time notifications
Add multiple evaluation agents for different roles
Implement multi-language support
Add candidate email automation

💡 Pro Tips

Better LinkedIn data: Ensure candidates provide complete LinkedIn URLs (not just usernames)
Consistent scoring: Run batches of similar roles together for normalized scoring
Cost optimization: Adjust Apify settings to fetch only essential data
Scale smartly: Process in batches of min 10-20 for optimal performance

⚠️ Important Notes

LinkedIn scraping respects Apify's rate limits
Scores are relative within each batch - don't compare across different job roles
The workflow handles both CSV and XLSX formats automatically
Error notifications help you catch issues before they cascade

📊 Expected Results

After implementation, expect:

Data-driven evaluation across candidates
Professional explanation for hiring decisions
Happy recruiters who can focus on human connection

Built with ❤️ by Elay Guez

Elay Guez

5 workflows

Nodes

n8n-nodes-base.formtrigger n8n-nodes-base.noop n8n-nodes-base.merge n8n-nodes-base.googlesheets n8n-nodes-base.googledrive n8n-nodes-base.extractfromfile n8n-nodes-base.code n8n-nodes-base.set

Complexity

advanced

Published 20 Sept 2025

Likes 0

View on n8n.io Download Workflow

Install path: /data/workflows/8775/8775.json

Share Your Workflow

Have a useful automation to share? Publish it and help the community.

Submit Your Template How to Submit

Related Workflows

Aggregate multi-source job boards to Supabase and Google Sheets

Stop manually checking dozens of career pages. This workflow runs every morning, hits the public APIs of 8+ ATS platforms and job boards, normalizes every listing into a single clean schema, and syncs everything to Supabase and Google Sheets deduplicated and ready to query. ## Who it's for Job seekers, recruiters, or career platforms that want a consolidated, up-to-date feed of openings from specific companies without scraping, without API keys, and without paying for a jobs aggregator. ## How it works 1. A Schedule Trigger fires daily at 8 AM IST 2. A Company List code node defines all sources grouped by ATS type (Greenhouse, Lever, Ashby, Workable, SmartRecruiters, RemoteOK, and board APIs like Remotive, Himalayas, Arbeitnow, Jobicy) 3. A Prepare Request node builds the correct API URL and headers for each source, including multi-page pagination for SmartRecruiters (up to 500 jobs via offset) and Himalayas (up to 500 via page param) 4. An HTTP Request node fetches all sources in batches of 5 5. A Parse + Enrich + Filter node normalizes all divergent JSON structures into a unified schema — resolving ISO country codes, Indian city detection, salary parsing across all formats, and domain-based filtering 6. Deduplicated results are upserted to a Supabase (Postgres) table and written to Google Sheets ## ATS platforms supported Greenhouse, Lever, Ashby, Workable, SmartRecruiters, RemoteOK, Remotive, Himalayas, Arbeitnow, Jobicy ## Normalized output schema `job_id`, `title`, `company`, `location`, `country`, `salary`, `job_type`, `apply_url`, `posted_at`, `source_ats` ## Setup 1. Open the **Company List** node and edit the `sources` array — add or remove companies and their ATS slugs 2. Update `ALLOWED_DOMAINS` in the Parse node to filter by location or job type relevant to you 3. Add your **Supabase** credentials in the Postgres node and confirm your table name and schema match the output fields 4. Connect your **Google Sheets** credentials and set the target spreadsheet and sheet ID 5. (Optional) Adjust pagination limits per source in the Prepare Request node ## Requirements - Self-hosted or cloud n8n instance - Supabase project with a jobs table - Google Sheets with headers matching the normalized schema - No external API keys required — all sources use public endpoints

View

Check and deactivate broken job apply URLs with Postgres and Google Sheets

Keep your job listings database clean without manual checks. Every three days, this workflow fetches all active jobs from your Postgres database, runs each application URL through a validation check, identifies dead links via HTTP status codes and soft-404 redirect detection, then marks failed entries as inactive in both Supabase and Google Sheets simultaneously. ## Who it's for Teams running a job aggregator, career platform, or internal hiring tracker who store job listings in Postgres and want stale or broken apply links removed automatically — without waiting for user reports. ## How it works 1. A Schedule Trigger fires every 3 days 2. All active jobs are fetched from your Postgres (Supabase) database via a SQL query 3. A Prepare URLs node filters out any rows with missing, malformed, or non-HTTP URLs before they're checked 4. An HTTP Request node sends a HEAD request to each `apply_url` 5. A Find Dead Jobs code node analyzes each response and flags a job as dead if: - Status code is `404` or `410` - DNS resolution fails (`ENOTFOUND`) - Connection is refused (`ECONNREFUSED`) - A `301`/`302`/`307` redirect points to a different path — indicating the job was removed and the ATS is silently redirecting (soft-404 detection) 6. If dead jobs are found, an IF node routes them to both update nodes in parallel: - **Supabase (Postgres)** — status set to `inactive` via parameterized SQL - **Google Sheets** — row updated to reflect the new status 7. If no dead jobs are detected, the workflow exits cleanly with no writes ## Setup 1. Connect your **Postgres** credentials and confirm the query in the Fetch Active Jobs node matches your table and column names (`apply_url`, `job_hash`, `job_title`) 2. Connect your **Google Sheets** credentials and set the Resource ID and Sheet Name in the Mark Inactive node 3. Confirm the `inactive` status value in the Postgres update query matches what your app expects 4. (Optional) Adjust the soft-404 redirect detection logic in the Find Dead Jobs node if your ATS platforms use non-standard redirect patterns ## Database columns expected `job_hash` (unique identifier), `apply_url`, `job_title`, `status` ## Requirements - Self-hosted or cloud n8n instance - Supabase (or any Postgres-compatible) database with an active jobs table - Google Sheets with a matching jobs log

View

Screen resumes with GPT-4 Turbo from Gmail and send shortlisted to Slack

# AI Resume Screening Workflow using Gmail, OpenAI (GPT-4 Turbo) & Slack This workflow automatically fetches resumes from Gmail, extracts text from PDF attachments, analyzes candidate profiles using AI, and shortlists qualified candidates by sending their details to Slack. It significantly reduces manual effort in resume screening and speeds up hiring decisions. ### Quick Implementation Steps 1. Connect your Gmail account in n8n 2. Connect your OpenAI API (GPT-4 Turbo) 3. Connect your Slack account 4. Ensure resumes are received as PDF attachments in Gmail 5. Update job requirements inside the AI Resume Analyzer node 6. Run the workflow manually or activate it ## What It Does This workflow automates the resume screening process by integrating Gmail, AI, and Slack into a seamless pipeline. It starts by fetching emails containing resume attachments, then processes and validates these files to ensure only PDF resumes are analyzed. Once validated, the workflow extracts text from each resume and sends it to an AI model. The AI evaluates candidate details such as skills, experience, and contact information, and calculates a match score based on predefined job requirements. Finally, the workflow determines whether a candidate should be shortlisted or rejected. Shortlisted candidates are instantly shared on Slack with complete details, enabling faster hiring decisions and reducing manual screening effort. ## Who’s It For - HR teams and recruiters - Hiring managers - Startups handling large volumes of applications - Recruitment agencies - Businesses looking to automate candidate screening ## Requirements to Use This Workflow - n8n (self-hosted or cloud) - Gmail account (OAuth connected) - OpenAI API key (GPT-4 Turbo access) - Slack account (with channel access) - Resumes in PDF format - Basic understanding of n8n workflows ## How It Works & Set Up ### Setup Instructions 1. **Manual Trigger Setup** Use the *Start Workflow Manually* node to initiate the workflow. 2. **Gmail Configuration** Connect your Gmail account in the *Fetch Gmail* node. Ensure emails contain resume attachments. 3. **Attachment Processing** The workflow formats and checks attachments. Only emails with attachments proceed further. 4. **PDF Validation** Ensure resumes are in PDF format. Non-PDF files are ignored automatically. 5. **Text Extraction** The *Extract Resume Text* node converts PDF resumes into plain text. 6. **AI Configuration** Add your OpenAI credentials in the *OpenAI Chat Model* node. Customize job requirements inside the *AI Resume Analyzer* node prompt. 7. **Evaluation Logic** Candidates are scored based on: - Skills (50%) - Experience (30%) - Location (20%) 8. **Output Handling** If score ≥ 70 → Sent to Slack as shortlisted. Else → Marked as rejected 9. **Activate Workflow** Run manually or activate for continuous use. ## How To Customize Nodes - **AI Resume Analyzer** - Modify job requirements (skills, experience, location). Adjust scoring logic or threshold - **Evaluate Candidate Score** - Change minimum score (default: 70) - **Slack Node** - Customize message format. Change Slack channel - **Gmail Node** - Add filters (labels, subjects, senders) - **Extract Resume Text** - Extend for other file formats if needed ## 1.8 Add-ons (Enhancements) - Store shortlisted candidates in Google Sheets or database - Send rejection emails automatically - Add ATS (Applicant Tracking System) integration - Include resume ranking system - Add duplicate candidate detection - Enable multi-role job matching ## 1.9 Use Case Examples - Automating resume screening for tech hiring - Filtering candidates for specific skill sets (e.g., React Native developers) - Handling bulk job applications efficiently - Shortlisting candidates for remote roles - Pre-screening candidates before interviews There can be many more use cases depending on hiring workflows and business needs. ## 1.10 Troubleshooting Guide | Issue | Possible Cause | Solution | |---------------------------|-----------------------------------------------------|------------------------------------------------------------| | No emails fetched | Gmail not connected or incorrect filters | Reconnect Gmail and check filters | | Attachments not processed | Emails don’t contain attachments | Ensure resumes are attached | | PDF not detected | Incorrect file type | Ensure resumes are in PDF format | | AI response parsing error | Invalid JSON from AI | Check prompt and enforce strict JSON | | No Slack message | Slack credentials or channel issue | Verify Slack connection and channel ID | | Low-quality matches | Incorrect job requirements | Update AI prompt with accurate criteria | ## Need Help? If you need assistance setting up this workflow or customizing it for your business needs, we’re here to help. Whether you want to: - Extend this workflow with advanced features - Integrate it with your existing systems - Build similar automation solutions Feel free to reach out to our [n8n workflow development](https://www.weblineindia.com/n8n-automation/) experts at **WeblineIndia** for expert support and tailored workflow development. 👉 Let us help you automate smarter and scale faster.

View

Need Custom Automation?

Get help designing a custom n8n workflow that connects your stack and fits your process.

Parse & evaluate HR candidates with GPT-4.1 and LinkedIn data in CSV/XLSX

Workflow preview

Important notice

Overview

AI-Powered HR Candidate Evaluation Agent with LinkedIn Data Enrichment in CSV/XLSX Format

🎯 Overview

👥 Who's it for?

⚡ How it works

🛠️ Setup Instructions

Requirements:

Step-by-step:

📋 Input File Format

🎨 Customization Options

Easy tweaks:

Advanced modifications:

💡 Pro Tips

⚠️ Important Notes

📊 Expected Results

Built with ❤️ by Elay Guez