CareerGPT AI-Powered Career Platform Development by Mohid NaghmanCareerGPT AI-Powered Career Platform Development by Mohid Naghman

CareerGPT AI-Powered Career Platform Development

Mohid Naghman

Completed work

AI Agent Developer

AI Engineer

Streamlit

Human Resources

🚀 CareerGPT - Your AI-Powered Career Intelligence Platform

CareerGPT is a sophisticated, multi-agent AI assistant designed to provide comprehensive career guidance. Built with a stateful LangGraph backend and a modern Streamlit frontend, this platform helps users explore careers, create personalized learning paths, receive in-depth resume analysis, and search for real-time job openings.

Demo: (#1 (comment))

✨ Core Features

🧠 Intelligent Agentic Workflow: A robust "hub and spoke" design using LangGraph. A central supervisor agent intelligently routes user requests to specialized agents for different tasks.

🗣️ Stateful & Context-Aware Conversations: The application remembers context across turns, enabling advanced features like follow-up questions about an analyzed resume.

📚 RAG-Powered Knowledge Base: The CareerAdvisor agent uses a local FAISS vector store to provide accurate, context-aware answers to general career questions.

🔎 Real-Time Job Search: The JobSearch agent integrates with the Tavily Search API to find, analyze, and summarize live job postings from the internet.

📄 Interactive Resume Coaching: Upload a resume for a detailed analysis, then ask specific questions and get contextual answers from the ResumeQAAgent.

✨ Polished, Modern UI: A responsive and intuitive user interface built with Streamlit and a custom, professional theme.

🛠️ Tech Stack

Category Technology / Service Frontend Streamlit Backend Python, LangChain, LangGraph LLMs Groq (Llama 3), Google AI (Gemini for Embeddings) Vector DB FAISS (Facebook AI Similarity Search) Tools Tavily Search API

📂 Project Structure

/CareerGPT/
├── agents/
│   ├── __init__.py
│   └── chains.py             # Logic for each agent's "brain"
├── data/
│   └── career_docs/          # Source PDFs for the RAG knowledge base
├── static/
│   └── style.css             # External CSS for the UI
├── utils/
│   ├── __init__.py
│   └── file_parser.py        # PDF and DOCX parsing utility
├── .env                      # Your secret API keys (Ignored by Git)
├── .gitignore                # Specifies files/folders for Git to ignore
├── Graph_backend.py          # The core LangGraph agentic workflow
├── requirements.txt          # Project dependencies
├── setup.py                  # One-time script to build the vector store
├── streamlit_ui.py           # The main Streamlit application file
└── README.md                 # You are here!

⚙️ Setup and Installation

Follow these steps to get CareerGPT running on your local machine.

1. Clone the Repository

git clone https://github.com/your-username/CareerGPT.git
cd CareerGPT

2. Create and Activate a Virtual Environment

# Create the environment
python -m venv venv

# Activate it
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activate

3. Install Dependencies

Install all required packages using the requirements.txt file.

pip install -r requirements.txt

4. Configure Environment Variables

You will need API keys from the following services:

Groq (for fast LLM inference)

Google AI Studio (for text embeddings)

Tavily AI (for the job search tool)

Create a file named .env in the project root. This file is included in .gitignore to keep your keys secure.

GROQ_API_KEY="your_groq_api_key_here"
GOOGLE_API_KEY="your_google_api_key_here"
TAVILY_API_KEY="your_tavily_api_key_here"

5. Build the Vector Store

Before running the app, you need to build the local FAISS index from the documents in /data/career_docs/.

python setup.py

This will create a faiss_index folder in your project root.

6. Run the Application

You're all set! Launch the Streamlit app:

streamlit run streamlit_ui.py

Navigate to http://localhost:8501 in your browser.

🚀 Future Enhancements

This project has a strong foundation with many possibilities for future development:

Interview Coach Agent: A new agent to conduct mock interviews and provide feedback.

Persistent User Sessions: Integrate a database (like SQLite) to save and load conversations.

Advanced RAG: Implement more sophisticated retrieval strategies like re-ranking.

Long-Term Memory: Use a vector store to give the chatbot long-term memory about the user's career goals and progress.

Like this project