From "Great!" to Real Learning: Building an AI Interview Teacher with Claude Code

ChatGPT Always Says "Great!" But I Remember Nothing — Built a Real Interview Teacher with Claude Code

A Universal Pain Point

Have you ever experienced this when using ChatGPT / Claude to prepare for interviews:

• You give an obviously flawed answer, and it replies "Great answer!" then moves straight to the next question
• Each time you start a new conversation, it doesn't remember what you learned yesterday, forcing you to reintroduce yourself
• You memorize a bunch of canned answers, but don't know when to review them, and forget everything after two days
• It explains concepts to you, you feel like you understand after reading, but when facing an interview, you can't articulate them

The root cause is simple: ChatGPT is an attentive chat companion, not a strict teacher.

Its operating principle is "make the user happy," not "make the user learn." It praises whatever you say because it fears you'll become unhappy.

Is There a Way to Make AI a Strict Teacher?

Yes. I achieved this using Claude Code (Anthropic's CLI programming tool).

Claude Code has a unique capability: Rules Files (CLAUDE.md) + Persistent Memory System. In simple terms, you can "program" the AI's behavior by writing rules files, and these rules are automatically loaded in every conversation.

I applied this capability to interview preparation.

What I Did

I created a set of configuration files, with two core components:

1. Strict Learning Process (rules/interview.md)

Rules:
1. Mix questions from different modules (3-4 areas, not just one direction)
2. Correct answer → Brief confirmation + Link to related knowledge → Next question
3. Wrong answer → Must explain (include reference URL) → Give verification question → Only proceed after passing
4. No skipping steps; cannot jump straight to the next question after answering

Compared to regular AI: Rules are hardcoded, not decided by the AI itself. It can't get lazy and say "Great" because the rules file forces it to complete the explanation → verification process.

2. Spaced Repetition Review Scheduling (knowledge/interview_tracker.md)

Knowledge Point	First Learned	D+1	D+2	D+4	D+7	D+15	Status
Smart Pointers	5/29	✅	✅				Mastered
Virtual Table	5/29	❌					Weak

At the start of each session, Claude automatically checks which knowledge points are due for review and prioritizes review questions. You don't need to remember when to review yourself.

3. Daily Wrap-up Verification

When you say "That's it for today," it triggers the wrap-up process:

1. Claude provides a summary outline of today's learning
2. You recall and provide your understanding based on the outline
3. Claude judges whether you truly understand or are just reciting answers
4. Updates the review schedule

This is the AI version of the Feynman Technique — not just feeling like you understand, but being able to explain it, with AI as the judge.

Feature Comparison

Capability	ChatGPT Interview Prep	Anki	NotebookLM	SaaS Platforms	This Project
Cross-session Memory	No	Deck-level only	No	Partial	Yes
Spaced Repetition	No	Yes	No	No	Yes
Wrong→Explain→Verify	No	No (just gives answer)	No	Partial	Yes
Native Question Generation from Project Code	Manual code pasting required	N/A	File upload required	Not supported	Native Support
Additional Costs	None	None	None	$29-300/session	No additional costs

Core difference: Other tools either have memory without understanding (Anki), or understanding without memory (ChatGPT), or lack both (SaaS). Claude Code's rules files + memory system is the only solution that combines both.

The Biggest Killer Feature: Starting from Your Project Code

All interview tools on the market follow the "I ask, you answer" model.

This project is different: Asks follow-up questions based on your own code.

For example, my project has a BoundedQueue (bounded queue) using two condition_variable. During interview preparation, Claude won't ask you "Please explain condition_variable," but rather:

"Your BoundedQueue uses two condition_variables (not_full_ and not_empty_). Could you use just one?"

Such questions force you to think about principles from code you've actually written, giving knowledge an anchor point rather than empty memorization of canned answers.

Quick Start

1. Fork the Repository

gh repo fork happiness-cheng/ai-interview-engine --clone

2. Open with Claude Code

cd ai-interview-engine
claude

3. Say Your First Words

I'm preparing for C++ backend development interviews, aiming for internship positions at top companies. Start reviewing.

That's it. Claude will read the rules files and begin the first round of questions.

4. Switch to Your Own Tech Stack

The repository includes sample question banks for C++ backend. If you're using Java / Go / Frontend, replace with the blank template:

knowledge/TEMPLATE.md → Copy as interview_tracker.md → Fill in your knowledge points

My User Experience

I'm a sophomore student preparing for the 2026 fall recruitment internship. Using this system for C++ backend interview preparation, the most noticeable benefits are:

1. No more anxiety — No need to plan "what to review today" myself; the system automatically schedules based on the forgetting curve
2. Actually remember — The wrong→explain→verify process forces genuine understanding, not just fleeting impressions
3. Confident about projects — Follow-up questions based on your own code let you clearly explain the "why" behind every design decision in your project

This Is Not Just an Interview Tool

The underlying engine is generic: Question generation → Answer → Evaluation → Explanation → Verification → Forgetting curve scheduling.

Interviews are just one scenario. You can use it to prepare for:

• Technical interviews (currently supported)
• Career transition/job change review
• Final exams (same principle, different knowledge base)

Open Source Repository

GitHub: https://github.com/happiness-cheng/ai-interview-engine

If this helps you, give it a star. If you have suggestions for improvement, feel free to open an issue.