AI-Powered HR Candidate Evaluation using GPT-4 and LinkedIn Data - n8n Workflow

HR Professionals drowning in manual candidate review.
Technical recruiters needing standardized, data-backed scoring.
Automation specialists looking for advanced n8n templates integrating AI agents and web scraping.
Teams requiring automated data enrichment (like LinkedIn analysis) before evaluation.

This comprehensive n8n workflow solves the tedious and subjective problem of evaluating large batches of HR candidates. By leveraging the power of a custom AI Agent (powered by GPT-4.1), the system not only standardizes the scoring process (0-100) but also enriches the data by scraping recent LinkedIn posts using Apify. The process starts with a simple file upload (CSV or XLSX). The n8n workflow then handles data conversion, storage, AI evaluation, and sheet formatting (including RTL support, column resizing, and sorting by score). This n8n automation is an excellent example of using specialized n8n node types to achieve complex, multi-system integration, drastically reducing evaluation time and improving consistency. This n8n workflow saves significant time, transforming hours of manual work into a scalable, automated process.

The entire process is initiated by an n8n trigger when a user submits a CSV or XLSX file via the built-in form trigger node.

1. Ingestion and Setup: The 'On form submission' n8n trigger captures the file, which is temporarily uploaded to Google Drive, downloaded, and then used to create a new Google Sheet document. A 'Switch' n8n node directs the binary data to the correct parser ('Extract from CSV' or 'Extract from XLSX').


2. Data Preparation: The extracted candidate data is appended to the new Google Sheet using a specific n8n node. Cleanup is performed by moving the original file to a temporary folder in Google Drive. Two subsequent n8n node operations prepare the new 'Evaluation Score' and 'Score Explanation' columns.


3. AI Orchestration: The 'Read Sheet' n8n node pulls all candidate rows, feeding the data into the 'AI Agent' n8n node. This agent, using the 'GPT-4.1' model, is instructed via a detailed system prompt to perform a critical sequence of tasks for each candidate: 1) Use the 'Run an Actor in Apify' n8n tool to fetch up to 3 recent LinkedIn posts using the profile URL. 2) Use the 'Get dataset items in Apify' n8n tool to retrieve the results. 3) Calculate a score (0-100) based on form answers and LinkedIn insights. 4) Use the 'Update row in sheet in Google Sheets' n8n tool to write the 'Evaluation Score' and 'Score Explanation' (in Hebrew) back to the correct row number.


4. Final Formatting: Once all candidates are processed, the n8n workflow uses an HTTP Request n8n node to interact directly with the Google Sheets API. This step applies advanced formatting, including enabling Right-to-Left (RTL) mode, setting up filters, bolding headers, and crucial automatic sorting of all candidates by the 'Evaluation Score' in descending order.


5. Error Handling: An 'Error Trigger' n8n node is configured to send an alert via Gmail if any unexpected failure occurs during the n8n workflow execution.

To deploy this powerful n8n workflow, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON data and paste it into your n8n instance using the 'New' -> 'Import from JSON' function.


2. Credentials Setup: This n8n template requires three primary credentials:

1. Activate the Trigger: The 'On form submission' n8n trigger node is automatically set up with a webhook ID. Activate the workflow to get the unique form URL where users can upload their CSV/XLSX files.


2. Testing: Run the n8n workflow manually with a test CSV file containing candidate names and LinkedIn URLs to verify data enrichment and AI scoring.

On form submission (n8n trigger): The starting point, providing a web form for users to upload candidate data files (CSV, XLSX, XLS).
AI Agent (Langchain Agent n8n node): The core intelligence, utilizing GPT-4.1. This n8n node orchestrates the entire evaluation logic, dynamically calling Apify tools for LinkedIn data and the Google Sheets Tool to update the 'Evaluation Score' and 'Score Explanation' fields.
GPT-4.1 (LLM Chat OpenAI n8n node): The specific LLM powering the AI Agent, configured for advanced reasoning and instruction following, including multilingual output (Hebrew explanations).
Run an Actor in Apify / Get dataset items in Apify (Apify Tool n8n nodes): Tools specifically integrated within the AI Agent's capabilities to execute the mandatory LinkedIn web scraping step for data enrichment.
Convert to Google Sheet (Google Sheets n8n node): Used to take the parsed CSV/XLSX data and append it as new rows into the dynamically created Google Sheet.
Apply Formatting (HTTP Request n8n node): A powerful n8n node that uses the Google Sheets API to apply complex, non-standard actions like RTL orientation, header styling, dynamic column resizing, and sorting the entire dataset by the 'Evaluation Score' column descending.