🎯 Interview Trainer Agent

AI-Powered Interview Preparation System | Edunet Foundation × IBM Internship

---

Project Overview

The Interview Trainer Agent is an intelligent, AI-powered application built using IBM Granite models via IBM watsonx.ai and LangFlow. It helps job seekers prepare for interviews by generating personalized, role-specific interview questions, model answers, and improvement tips based on their job role, experience level, and skills.

This project was developed as part of the Edunet Foundation × IBM SkillsBuild Internship Program.

---

Features

• Conversational Chat Interface — Interact naturally to get interview questions
• Resume/Document Upload — Upload your resume (PDF/TXT/DOCX) for personalized preparation
• Role-Specific Questions — Tailored technical and behavioral questions for any job role
• Model Answers — Detailed model answers for every question generated
• Improvement Tips — Actionable tips to improve your interview performance
• Quick Prompts — One-click preparation for popular job roles
• Powered by IBM Granite — Enterprise-grade AI via IBM watsonx.ai

---

Tech Stack

Technology	Purpose
IBM Granite (ibm/granite-4-h-small)	Core AI model for question generation
IBM watsonx.ai	Model deployment and AI governance
IBM Cloud	Cloud infrastructure
LangFlow	Visual agent workflow orchestration
Streamlit	Frontend chat interface
Python	Backend application logic

---

Architecture

User Input (Chat / Resume Upload)
            ↓
    Streamlit Frontend (app.py)
            ↓
    LangFlow API (localhost:7860)
            ↓
  ┌─────────────────────────────┐
  │     LangFlow Agent Flow      │
  │  ┌──────────┐  ┌─────────┐  │
  │  │Chat Input│  │ Prompt  │  │
  │  └────┬─────┘  │Template │  │
  │       │        └────┬────┘  │
  │       └──────┬───────┘      │
  │          ┌───▼────┐         │
  │          │ Agent  │         │
  │          │IBM     │         │
  │          │Granite │         │
  │          └───┬────┘         │
  │          ┌───▼────┐         │
  │          │ Chat   │         │
  │          │ Output │         │
  │          └────────┘         │
  └─────────────────────────────┘
            ↓
    Interview Questions + Model Answers + Tips

---

Installation & Setup

Prerequisites

• Python 3.10 or above
• pip or uv package manager
• IBM Cloud account (free/Lite tier)
• IBM watsonx.ai project with API key

Step 1 — Clone the Repository

git clone https://github.com/saibhargavi-rapolu/Interview-Trainer-Agent.git
cd Interview-Trainer-Agent

Step 2 — Install Dependencies

pip install -r requirements.txt

Step 3 — Install and Start LangFlow

pip install uv
uv venv langflow
langflow\Scripts\activate       # Windows
uv pip install langflow
uv run langflow run

LangFlow will start at: http://localhost:7860

Step 4 — Configure IBM watsonx.ai in LangFlow

1. Open LangFlow at http://localhost:7860
2. Create a new blank flow
3. Add components: Chat Input → Prompt Template → Agent (IBM Granite) → Chat Output
4. In the Agent component, enter your:
   • IBM Cloud API Key
   • watsonx Project ID
   • Endpoint: https://us-south.ml.cloud.ibm.com
5. Select model: ibm/granite-4-h-small

Step 5 — Run the Streamlit App

streamlit run app.py

Open your browser at: http://localhost:8501

---

---

Environment Variables

Update these values in app.py:

LANGFLOW_URL = "http://127.0.0.1:7860"
FLOW_ID = "your-langflow-flow-id"
LANGFLOW_API_KEY = "your-langflow-api-key"

---

Requirements

streamlit
requests
langflow

---

How to Use

1. Start LangFlow — Run uv run langflow run in terminal
2. Start the app — Run streamlit run app.py in another terminal
3. Open browser — Go to http://localhost:8501
4. Chat or Upload — Type your job role or upload your resume
5. Get Questions — Receive personalized interview questions and answers instantly

Example Prompts

"I am applying for a Python Developer role with 1 year of experience."
"I am a Data Scientist with 3 years experience in ML and Python."
"I am a fresher applying for a Web Developer position."

---

LangFlow Agent Prompt

The agent uses this optimized system prompt:

You are an expert Interview Trainer Agent powered by IBM Granite.
When a user provides their job role and experience level, respond with:

##  Interview Preparation Guide
###  Technical Questions (5 questions)
###  Behavioral/HR Questions (3 questions)
###  Improvement Tips (3 tips)
###  Key Topics to Study

---

Key Highlights

• Role-Specific Precision — Questions tailored to exact job title and experience
• RAG-Powered — Retrieves real interview patterns in real-time
• Dual Assessment — Covers both technical skills and behavioral/soft skills
• Model Answers — Not just questions, but how to answer them effectively
• Resume-Aware — Analyzes uploaded resume for personalized preparation
• IBM Enterprise-Grade — Powered by IBM Granite for reliable, governed AI

---

Future Scope

1. Voice Interview Simulation — Practice speaking answers with AI feedback
2. Company-Specific Preparation — Questions based on specific company patterns
3. Real-Time Performance Analytics — Dashboard showing scores and improvement trends
4. Multilingual Support — Interview prep in Hindi, Telugu, and other languages

---

🌐 Live Demo

**Click here to try the app : https://interview-trainer-ibm.streamlit.app **

Author

Rapolu Sai Bhargavi Edunet Foundation × IBM SkillsBuild Internship

---

License

This project is licensed under the MIT License — see the LICENSE file for details.

---

Acknowledgements

• Edunet Foundation — For providing this internship opportunity
• IBM — For IBM Granite models and watsonx.ai platform
• LangFlow — For the visual agent workflow builder