Tech Interview Generator

Generate high-quality technical interview questions instantly using AI. Perfect for hiring managers, interviewers, and anyone preparing for technical interviews.

Built with DSPy - a framework for programming (not prompting) language models.

Quick Start (2 minutes)

# 1. Clone and setup
git clone https://github.com/anyakonovalenko/tech-interview-generator.git
cd tech-interview-generator
bash setup.sh

# 2. Add your OpenAI API key to .env file
# Get one at: https://platform.openai.com/api-keys

# 3. Generate questions!
python generate_question.py

Usage

Interactive Mode (Easiest)

python generate_question.py

Follow the prompts to select:

• Question type (Coding, ML Theory, ML Practical)
• Topic (e.g., "binary trees", "neural networks")
• Difficulty (easy, medium, hard)
• Number of follow-up questions

Quick Mode

# Generate question on a specific topic
python generate_question.py "dynamic programming"
python generate_question.py "convolutional neural networks"
python generate_question.py "graph algorithms"

Features

• Multiple Question Types: Coding/algorithms, ML theory, ML practical
• Smart Difficulty Levels: Easy, medium, hard with automatic validation
• Follow-up Questions: Get 1-4 related follow-ups automatically
• Instant Generation: Powered by GPT-4o-mini for fast results
• No Prompt Engineering: Built with DSPy for robust, reliable outputs

Example Output

==================================================
       TECH INTERVIEW QUESTION GENERATOR
==================================================

Topic: binary search trees
Type: coding
Difficulty: medium

------------------------------------------------------------
Design and implement a function to find the kth smallest
element in a Binary Search Tree (BST).

Input: Root node of BST and integer k
Output: The kth smallest element

Example:
    BST:     5
           /   \
          3     7
         / \   / \
        2   4 6   8

    k = 3 → Output: 4

Constraints:
- Time complexity should be O(k) on average
- Space complexity should be O(h) where h is tree height
------------------------------------------------------------

FOLLOW-UP QUESTIONS:
1. How would you modify your solution to handle frequent
   queries for different k values on the same BST?
2. What changes are needed if the BST is modified frequently
   with insertions and deletions?
3. Can you solve this without using extra space?

Use Cases

• Hiring Managers: Generate diverse questions for technical interviews
• Interview Prep: Practice with realistic interview questions
• Educators: Create assignments and exam questions
• Self-Study: Test your knowledge on specific topics

For Developers: Learning DSPy

This project is built with DSPy, a framework for programming (not prompting) language models. If you want to understand or extend the code:

Project Structure:

tech-interview-generator/
├── generate_question.py     # Main CLI tool (start here!)
├── setup.sh                 # Quick setup script
├── src/interview_generator/
│   ├── signatures.py        # DSPy task declarations
│   └── modules.py           # Question generation logic
└── examples/                # Learning examples
    ├── basic_usage.py
    ├── advanced_pipeline.py
    └── understanding_dspy.py

Run the examples to learn DSPy:

python examples/understanding_dspy.py  # Learn DSPy concepts
python examples/basic_usage.py         # Simple usage
python examples/advanced_pipeline.py   # Advanced patterns

Resources:

• DSPy Documentation
• DSPy GitHub
• Why DSPy? - No more prompt engineering!

Contributing

This is a learning project! Contributions are welcome:

• Add new question types
• Improve existing modules
• Add optimization examples
• Enhance documentation

License

MIT License - feel free to use this project for learning and building your own tools.

Acknowledgments

• Built with DSPy by Stanford NLP
• Inspired by the need for better technical interview preparation tools