AI-Powered Excel Mock Interviewer

📖 Project Overview

The AI-Powered Excel Mock Interviewer is an automated system designed to assess candidates' Microsoft Excel skills efficiently and consistently. As our company rapidly expands its Finance, Operations, and Data Analytics teams, we need a scalable and intelligent solution to evaluate technical proficiency without relying on time-consuming and inconsistent manual interviews.

This project simulates a real-world interview experience by dynamically generating Excel-related questions, evaluating responses, providing feedback, and adjusting difficulty based on performance. At the end of each session, candidates receive a comprehensive report summarizing their strengths and areas for improvement.

---

🎯 Problem Statement

• Manual Excel interviews are resource-intensive and inconsistent.
• A scalable, automated solution is needed to screen candidates faster.
• The system must provide a structured interview experience with intelligent evaluation.
• It must handle sessions dynamically, provide feedback, and generate performance reports.

---

🚀 Features

✅ Structured multi-turn interview with a conversational flow
✅ Domain-specific questions tailored to Data Analysis, Finance, or Operations
✅ Real-time answer evaluation using a language model (LLM)
✅ Adaptive difficulty based on candidate performance
✅ Detailed feedback after each question
✅ Summary report with scoring and explanations
✅ Resume session management and dynamic state tracking
✅ Support for next candidate interviews

---

🛠 Technology Stack

• Ollama API – LLM integration for question generation and answer evaluation
• FastAPI – Backend framework for managing sessions and API endpoints
• Streamlit – Frontend interface for candidate interaction and feedback
• Python – Programming language powering the entire system

---

📂 Project Structure

excel_interviewer_project/
├── backend/
│   ├── __init__.py          # Marks backend as a Python package
│   ├── main.py              # FastAPI app handling API endpoints
│   └── interview_logic.py   # Core logic for question generation, scoring, and saving reports
├── frontend/
│   └── app.py              # Streamlit app for UI and interaction
├── report/                 # Folder where transcripts are saved
├── requirements.txt        # Python dependencies
└── README.md              # Project documentation

---

📥 Installation Instructions

1. Clone the repository

git clone <your-repo-url>
cd excel_interviewer_project

2. Install dependencies

pip install -r requirements.txt

Make sure you have Python 3.8+ installed.

3. Run the backend server

cd backend
uvicorn main:app --reload

The backend will be available at http://127.0.0.1:8000.

4. Run the frontend app

cd frontend
streamlit run app.py

The frontend will open in your browser.

---

⚙ Environment Variables

You can configure the following environment variables:

• OLLAMA_URL – URL of the LLM API (default: http://localhost:11434/api/generate)
• OLLAMA_MODEL – Model to use for question generation (default: mistral:7b)
• MAX_LLM_RETRIES – Number of retries when contacting the API (default: 2)
• API_URL – URL of the FastAPI backend (default: http://127.0.0.1:8000)

You can create a .env file or set them in your environment before running the apps.

---

📊 Example Prompts

Question Generation Prompt:

You are a knowledgeable Excel interviewer specialized in Data Analysis. The candidate has answered 3 interview questions with 2 correct and 1 wrong answers so far. Ask the next question at difficulty level 5 (0=easy, 10=hard). Provide a clear, practical Excel interview question and an ideal answer as reference.

Evaluation Prompt:

You are an expert Excel interviewer evaluator. Given the interview question: "How do you sum values in column B where column A equals 'X'?" and the candidate's answer, evaluate correctness and provide a numeric score between 0 and 1.

---

📈 Future Improvements

✔ Collect transcripts to refine scoring and question prompts
✔ Incorporate candidate feedback for better usability
✔ Integrate with cloud storage for scalability
✔ Fine-tune models based on real-world data
✔ Add analytics and reporting dashboards

---

📜 License

This project is for educational purposes and internal use only.

---

🙌 Acknowledgments

• Powered by Python, FastAPI, Streamlit, and Ollama.
• Inspired by the need for scalable and intelligent recruitment solutions.

If you encounter any issues or have suggestions, feel free to open an issue or contribute via pull requests!

---

Need help with deployment or additional features?

Let me know if you want me to: ✔ Tailor it further for deployment environments (Heroku, Streamlit Cloud, etc.)
✔ Add sample .env file templates or detailed API specs
✔ Help you push this to GitHub with commit messages and organization guidelines 📂📑✅