How to Build an AI Resume Parser

Project Access Link

https://editor.momen.app/tool/wxjD4gygY2o/WEB?code=6ltP5UlOVRcVW 

Introduction

• Goal: Automatically extract key candidate information from uploaded resume files using AI.
• Applicable Scenario: Recruitment platforms, HR management systems, or automated candidate screening tools.
• Core Logic: User uploads a file to a Page Variable -> Button click triggers an Actionflow -> Actionflow saves the file, invokes the AI Agent, and updates the record with parsed data.

Steps

Data Storage

First, we need a table to store both the original resume files and the parsed information.

• Data Model: Create a table named candidate with the following fields:

Logic & State Configuration

AI Agent Configuration

1. Navigate to the AI tab and create a new agent named agent_resume_parser.
2. Inputs: Add an input named resume with the type File.
3. Prompt Template:
   • Role: You are a professional HR Resume Parser.
   • Goals: Read the content of the input resume file, accurately extract key fields, and convert them into a structured JSON format.
4. Outputs: Set the type to Structured. Define the following fields:
   • name (String): Candidate’s full name.
   • core_skills (String): Core hard skills.
   • years_of_experience (Integer): Total work years.

Actionflow Construction

Create an Actionflow named AI Resume Parser to handle the end-to-end logic.

• Trigger: Set an input parameter named resume (Type: File) at the Input node.
• Action Steps:
  1. save resume (Insert Data): Insert a new record into the candidate table. Bind the resume field to the Actionflow input resume.
  2. AI node (Start conversation): Call the agent_resume_parser. Bind its resume input to the Actionflow input resume.
  3. update candidate info (Update Data):
     • Filter: Use the ID from the save resume node (save resume.id).
     • Mapping: Map the AI Agent’s output fields (name, core_skills, years_of_experience) to the database columns.

UI Construction & Interaction

1. In the Page editor, select the current page from the Component Tree.

2. On the Right Sidebar, navigate to the Data tab.

3. Click + Add under Page variable.

4. Set the name to resume and the type to File.

5. Component Tree: Add a Button component to the canvas and set its content to “Get Started”.

6. Interaction Configuration: Select the button and open the Action panel on the right.

7. Upload File:

   • Add the “Upload file” action.
   • In the “After upload” setting, assign the FileInfo to the Page variable resume.

8. Run Actionflow:

   • Under Actions on success of the upload action, add the Actionflow action.
   • Select AI Resume Parser.
   • Bind the resume parameter to Page variable.resume.

Verification

1. Click the Preview button in the top right.
2. Click “Get Started”, select a resume PDF, and upload it.
3. Go to the Data Source -> Database tab in the editor.
4. Check the candidate table. You should see a new record where the full_name, core_skills, and years_of_experience have been automatically filled by the AI.