Python Screening Task 2 – AI Debugging Assistant

1. Task Objective

This task evaluates the ability to craft a clear, effective prompt that guides an AI assistant to help a student debug Python code without revealing the correct solution.

The AI must:

• Analyze a student’s buggy Python code
• Offer helpful hints, suggestions, or debugging strategies
• Avoid giving away the fixed code or complete solution

---

2. Design Choices

a) Why worded this way

• Used clear, action-oriented instructions (examine, explain, offer hints) to keep the AI focused on teaching, not fixing.
• Chose words like supportive and encouraging to set a student-friendly tone.

b) Avoiding solutions

• Explicitly stated: “Do not provide corrected code or a complete solution.”
• Encourages guiding questions and hints instead of spoon-feeding answers.

c) Student-friendly feedback

• Added phrases like gentle hints and encouraging to push the AI toward a mentor-like role, offering reassurance while challenging students to think.

---

3. Reasoning

a) Tone and Style

• Should be supportive, patient, and approachable — like a friendly tutor.
• Use simple, empathetic language to reduce frustration while staying professional.

b) Balance: Identifying Bugs vs. Guiding

1. Step 1: Point out likely problem areas.
2. Step 2: Explain why those issues matter.
3. Step 3: Provide hints or questions that nudge students toward the fix.

c) Adaptation for Learners

• Beginners:

  • Use simple language, break down steps clearly, and avoid jargon.
  • Provide direct but non-revealing hints.

• Advanced Learners:

  • Highlight deeper conceptual errors and edge cases.
  • Encourage debugging strategies (e.g., “What happens if you print this variable?”).

---

4. Setup Instructions

1. Clone this repository or download the files.
2. The main submission is in this README.md file.
3. Open and review the Design Choices and Reasoning sections.
4. Share the repo link or this file with the evaluation team as per submission instructions.

---