I Built a Resume Parser with ChatGPT, Python, and Zero Experience in NLP

Photo by Mr. Bochelly on Unsplash

I’m not a Natural Language Processing (NLP) expert.

In fact, before this project, I barely knew what tokenization or entity recognition meant.

But thanks to Python, some clever Googling, and ChatGPT acting like my 24/7 coding buddy, I managed to build a functional resume parser in just a couple of days.

Here’s how it happened — what I built, how I built it, and why you can (and should) build it too.

Photo by Mohammad Rahmani on Unsplash

💡 The Idea

I was applying for internships and realized: every resume is different, and recruiters don’t have time to manually extract skills, experience, education, etc.

Wouldn’t it be cool to build a Python script that could read a PDF resume and extract the key info in seconds?

That’s what I set out to do.

🔧 Tools I Used

• Python
• PyMuPDF (fitz) — to extract text from PDF
• spaCy — for basic NLP tasks
• ChatGPT — for guidance, code explanations, and quick fixes
• re (regex) — to match patterns like email, phone numbers
• Streamlit (optional) — to make it a simple web app

Photo by AltumCode on Unsplash

🛠️ Step-by-Step Breakdown

1. Extract Text from PDF

import fitz  # PyMuPDF

def extract_text_from_pdf(pdf_path):
    doc = fitz.open(pdf_path)
    text = ""
    for page in doc:
        text += page.get_text()
    return text

ChatGPT helped me tweak this to handle multi-page resumes too.

2. Extract Name, Email, Phone

import re

def extract_email(text):
    email = re.findall(r'\S+@\S+', text)
    return email[0] if email else None

def extract_phone(text):
    phone = re.findall(r'\+?\d[\d\s.-]{8,}\d', text)
    return phone[0] if phone else None

Simple regex + GPT-powered pattern tweaks = good enough accuracy for most resumes.

Photo by Christopher Gower on Unsplash

3. Extract Skills (Using spaCy + a Skills List)

import spacy

nlp = spacy.load("en_core_web_sm")
skills_list = ["python", "machine learning", "sql", "javascript", "react", "excel"]

def extract_skills(text):
    text = text.lower()
    return [skill for skill in skills_list if skill in text]

Later, I asked GPT to help me auto-expand this list using job descriptions. Game changer.

4. Bonus: Use ChatGPT to Extract Experience Sections

Here’s the fun part. I sent the raw PDF text to ChatGPT and asked:

“Can you extract the Experience section and turn it into structured data?”

And boom — it worked.

Get Rohansoni’s stories in your inbox

Join Medium for free to get updates from this writer.

Remember me for faster sign in

ChatGPT gave me back a clean JSON-style summary:

{
  "Company": "Google",
  "Role": "Software Intern",
  "Duration": "June 2023 - August 2023",
  "Description": "Worked on backend optimization for internal APIs"
}

Yes, seriously.

Photo by Brett Jordan on Unsplash

🚀 What I Learned

• Regex is your best friend (and sometimes your worst enemy).
• ChatGPT can debug, explain, and improve your code on the fly.
• You don’t need to be an NLP expert to start using NLP.

🎁 Bonus: Make It a Web App with Streamlit

import streamlit as st

uploaded_file = st.file_uploader("Upload Resume PDF")
if uploaded_file:
    text = extract_text_from_pdf(uploaded_file)
    st.write("Email:", extract_email(text))
    st.write("Phone:", extract_phone(text))
    st.write("Skills:", extract_skills(text))

Now it’s shareable with friends, recruiters, or even as a portfolio project.

Photo by Patrick Amoy on Unsplash

🧠 Final Thoughts

Was it perfect? No. Was it fun and surprisingly powerful? Absolutely.

Building this resume parser taught me that you don’t need to be a guru to build useful tools. You just need:

• Curiosity
• A goal
• And maybe a little help from ChatGPT

If you’re learning NLP, try this project. It’s beginner-friendly, resume-worthy, and genuinely helpful.

And if you want the full code, drop a comment 📈

A Message from AI Mind

Thanks for being a part of our community! Before you go:

• 👏 Clap for the story and follow the author 👉
• 📰 View more content in the AI Mind Publication
• 🧠 Improve your AI prompts effortlessly and FREE
• 🧰 Discover Intuitive AI Tools