---
title: "How to Teach Programming in the AI Era? Using LLMs as a Teachable Agent for Debugging (Extended Abstract)"
authors:
  - "Qianou Ma"
  - "Hua Shen"
  - "Ken Koedinger"
  - "Tongshuang Wu"
doi: "10.24963/ijcai.2025/1217"
url: "https://ijcai.org/proceedings/2025/1217"
pdf: "https://ijcai.org/proceedings/2025/1217.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "10922-10926"
track: "Sister Conferences Best Papers"
keywords:
  - "Sister Conferences Best Papers: Humans and AI"
  - "Sister Conferences Best Papers: Multidisciplinary Topics and Applications"
  - "Sister Conferences Best Papers: Agent-based and Multi-agent Systems"
---

# How to Teach Programming in the AI Era? Using LLMs as a Teachable Agent for Debugging (Extended Abstract)

Qianou Ma, Hua Shen, Ken Koedinger, Tongshuang Wu

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 10922-10926. DOI: [10.24963/ijcai.2025/1217](https://doi.org/10.24963/ijcai.2025/1217)

## Abstract

Large Language Models (LLMs) excel at generating content at impeccable speeds. However, they are imperfect and still make various mistakes. In Computer Science education, as LLMs are widely recognized as "AI pair programmers," it becomes increasingly important to train students on evaluating and debugging LLM-generated codes. In this work, we introduce HypoCompass, a novel system to facilitate deliberate practice on debugging, where human novices play the role of Teaching Assistants and help LLM-powered teachable agents debug code.
We enable effective task delegation between students and LLMs in this learning-by-teaching environment: students focus on hypothesizing the cause of code errors, while adjacent skills like code completion are offloaded to LLM-agents. Our evaluations demonstrate that HypoCompass generates high-quality training materials (e.g., bugs and fixes), outperforming human counterparts fourfold in efficiency, and significantly improves student performance on debugging by 12% in the pre-to-post test.