SCRIPT: Implementing an Intelligent Tutoring System for Programming in a German University Context

2026-04-21 · Source: cs.AI updates on arXiv.org · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Data Science & Analytics · Depth: Advanced, long

Summary

SCRIPT (Step-based Coding for Research and Intelligent Programming Tutoring) is a novel Intelligent Tutoring System (ITS) developed by researchers at Bielefeld University, Germany, to support advanced undergraduate computer science students learning Python for data science and machine learning. Unlike existing ITSs, SCRIPT specifically supports Python, targets advanced programming, and integrates generative models for hint mechanisms. It is designed as both a teaching and research platform, capable of recording fine-grained, keystroke-level data for A/B testing of learner and pedagogical models. Crucially, SCRIPT adheres to strict European regulatory requirements, including GDPR, the European AI Act, and the German Research Foundation's ethical framework, by using pseudonymous data, explicit consent for research data, and self-hosting a Llama-70b model to avoid commercial LLM interfaces.

Key takeaway

For AI Scientists and Machine Learning Engineers developing educational tools in regulated environments, SCRIPT demonstrates a robust approach to compliance and research integration. Your teams should prioritize modular architectures and self-hosted open-weight LLMs to navigate data protection regulations like GDPR and the European AI Act. Anticipate significant effort for ethical approvals and compliance procedures, as these can delay project rollout but ultimately improve system integrity and user trust.

Key insights

SCRIPT is a Python ITS designed for advanced learners, integrating LLMs and strict EU data privacy.

Principles

• ITSs bridge educational demand-supply gaps.
• Regulatory compliance is paramount for AIED systems.
• Modular architecture enhances ITS adaptability.

Method

SCRIPT employs a 4-model ITS approach (Pedagogical, Learner, Domain, UI) within a Dockerized web application. It uses FastAPI, MongoDB, Judge0 for secure code execution, and Ollama to host open-weight LLMs for hint generation.

In practice

• Use pseudonymous usernames for GDPR compliance.
• Self-host LLMs to control data privacy and reproducibility.
• Implement explicit, granular consent for research data.

Topics

• Intelligent Tutoring Systems
• Python Programming Education
• Large Language Models
• SCRIPT
• GDPR

Best for: AI Scientist, Machine Learning Engineer, Research Scientist

Related on AIssential

• SCRIPT: Implementing an Intelligent Tutoring System for Programming in a German University Context — A new Python ITS is being built to provide adaptable, AI-enhanced, and regulation-compliant programming education.
• The UK government’s best path is to stop treating this as a binary choice (“exception” vs “no exception”) and instead build a two-sided market architecture... — The UK is prioritizing a "licensing economy" for AI, focusing on transparency and infrastructure over broad copyright exceptions.
• How Software Engineering Students Use LLMs to Write Research Papers: An Experience Report — Students use LLMs across academic writing, balancing utility for clarity and organization with critical verification of AI-generated content.
• UK: ICO Report on Automated Decision-Making in Recruitment — AI tools in recruitment offer efficiency but face significant challenges with data protection regulations.
• A Student’s Response to Berkeley Law’s AI Ban — Transparent, guided AI integration in legal education fosters learning and accountability more effectively than outright bans.
• 95% of UK students now use AI and their experiences couldn't be more divided — Widespread AI adoption among UK students creates a critical gap in university support and raises concerns about skill development.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.AI updates on arXiv.org.