Skip to content

task.MD

Latest commit

 

History

History
111 lines (82 loc) · 4.14 KB

task.MD

File metadata and controls

111 lines (82 loc) · 4.14 KB
Raw

Assignment Title: Build a Simple Conversational AI Chatbot

Objective

To evaluate the candidate's ability to:

  1. Understand the basics of conversational AI and chatbot development.
  2. Use pre-trained models to build a functional chatbot.
  3. Demonstrate creativity and problem-solving skills in creating a simple, user-friendly chatbot interface.

Task Description

Create a Conversational AI Chatbot that can answer basic questions or have simple conversations with a user. The chatbot should have the following features:

  1. Input

    • Accept text input from the user (e.g., via a command-line interface, a simple web app, or a chat interface).

  2. Processing

    • Use a pre-trained conversational AI model (e.g., OpenAI’s GPT-3.5, Hugging Face models, or similar).
    • Ensure the chatbot responds coherently to user queries.

  3. Output

    • Display the chatbot’s response in real-time.

Functional Requirements

  1. The chatbot should be able to:

    • Greet the user (e.g., “Hello! How can I help you today?”).
    • Respond to simple general knowledge questions (e.g., “What’s the capital of France?”).
    • Handle casual conversation (e.g., “How’s the weather today?” or “Tell me a joke.”)

  2. The chatbot should handle basic edge cases:

    • If it doesn’t understand a query, respond with something like:
      “I’m sorry, I didn’t understand that. Could you please rephrase?”

Bonus Features (Optional)

  1. Predefined Intent Recognition

    • Recognize specific intents like:
      • Weather inquiries: “What’s the weather like today?”
      • Time-related queries: “What time is it?”
    • Respond accordingly.

  2. Web Interface

    • Create a simple web-based interface using a framework like Flask, Streamlit, or React.js.

  3. Personalization

    • Allow the user to provide their name and personalize responses (e.g., “Hi John, how can I assist you?”).

  4. Memory

    • Enable the chatbot to remember the context of the conversation within a session (e.g., “Can you tell me more about that?”).

Technical Requirements

  1. Programming Language

    • Use Python for development.

  2. Pre-Trained Model

    • Use a conversational AI model such as:
      • OpenAI's API (GPT-3.5 or similar).
      • Hugging Face’s transformers library with models like DialoGPT or T5.
      • Or any basic chatbot library like ChatterBot.

  3. Libraries/Tools

    • Suggested libraries: transformers, flask (for a web interface), nltk, or any other you’re comfortable with.

  4. Error Handling

    • Handle empty or invalid input gracefully.

  5. Documentation

    • Provide clear instructions in a README.md file on how to set up and run the chatbot.

Deliverables

  1. Source Code

    • A GitHub repository containing:
      • The Python code for the chatbot.
      • Any configuration or requirement files (e.g., requirements.txt).

  2. README.md

    • Clear setup instructions.
    • Examples of queries the chatbot can handle.

  3. Sample Output

    • Provide a few examples of conversations with the chatbot.

  4. Optional Deliverables

    • A link to the web-based interface (if implemented).

Evaluation Criteria

  1. Understanding of Conversational AI

    • Does the chatbot handle user queries logically and conversationally?

  2. Functionality

    • Does the chatbot work as described in the task requirements?

  3. Code Quality

    • Is the code clean, modular, and readable?
    • Are best practices followed for error handling and maintainability?

  4. Creativity

    • Have any bonus features been implemented, and are they functional?

  5. Documentation

    • Is the README.md file clear and helpful?

Tips for Candidates

  1. Start Simple

    • Focus on getting the chatbot to respond to a few basic queries before adding extra features.

  2. Use Online Resources

    • Look for tutorials or documentation for tools like Hugging Face or OpenAI’s API.

  3. Test as You Go

    • Regularly test the chatbot to ensure it behaves as expected.

  4. Add Comments

    • Write clear comments in your code to explain key sections.