How to Intermediate · 3 min read

How to build AI-powered resume parser

Q: How to build AI-powered resume parser

Use a large language model like gpt-4o to parse resumes by prompting it to extract structured fields such as name, skills, and experience. Send the resume text to client.chat.completions.create with a clear system prompt to guide the extraction.

Quick answer

Use a large language model like gpt-4o to parse resumes by prompting it to extract structured fields such as name, skills, and experience. Send the resume text to client.chat.completions.create with a clear system prompt to guide the extraction.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the OpenAI Python SDK and set your API key as an environment variable for secure access.

bash

pip install openai>=1.0

Step by step

This example shows how to send a resume text to gpt-4o and parse key fields like name, email, skills, and experience.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

resume_text = '''\nJohn Doe\nEmail: john.doe@example.com\nPhone: (555) 123-4567\nSkills: Python, JavaScript, Machine Learning\nExperience:\n- Software Engineer at TechCorp (2019-2023)\n- Data Analyst at DataWorks (2017-2019)\n'''

system_prompt = (
    "You are an expert resume parser. Extract the following fields from the resume text: "
    "Name, Email, Phone, Skills (as a list), Experience (as a list of job titles with dates). "
    "Return the result as a JSON object."
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": system_prompt},
        {"role": "user", "content": resume_text}
    ]
)

parsed_resume = response.choices[0].message.content
print(parsed_resume)

output

{
  "Name": "John Doe",
  "Email": "john.doe@example.com",
  "Phone": "(555) 123-4567",
  "Skills": ["Python", "JavaScript", "Machine Learning"],
  "Experience": [
    "Software Engineer at TechCorp (2019-2023)",
    "Data Analyst at DataWorks (2017-2019)"
  ]
}

Common variations

You can use asynchronous calls for better performance or switch to other models like claude-3-5-sonnet-20241022 for improved parsing accuracy. Streaming responses can be used for large resumes.

python

import os
import asyncio
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

async def parse_resume_async(resume_text: str):
    response = await client.chat.completions.acreate(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": "Extract name, email, phone, skills, and experience as JSON."},
            {"role": "user", "content": resume_text}
        ]
    )
    return response.choices[0].message.content

resume_text = "John Doe\nEmail: john.doe@example.com\nSkills: Python, JavaScript"

parsed = asyncio.run(parse_resume_async(resume_text))
print(parsed)

output

{
  "Name": "John Doe",
  "Email": "john.doe@example.com",
  "Skills": ["Python", "JavaScript"]
}

Troubleshooting

If the output is not valid JSON, add instructions to strictly format the response as JSON.
If the model misses fields, increase prompt clarity or use few-shot examples.
For rate limits, implement exponential backoff retries.

✅

Key Takeaways

Use clear system prompts to guide the LLM to extract structured resume data.
The gpt-4o model is effective for parsing resumes into JSON format.
Async calls and streaming can improve performance for large or multiple resumes.

Verified 2026-04 · gpt-4o, claude-3-5-sonnet-20241022

Verify ↗