AI Pseudocode tutor

Inspiration

As a Cambridge International A Level Computer Science student studying the 9618 syllabus, I know that pseudocode can be challenging to learn independently. Students often understand the general programming concepts but struggle with pseudocode syntax, exam-style questions, and understanding why a particular solution works.

I wanted to create a tool that could act as an accessible learning companion rather than simply providing answers. The idea behind AI Pseudocode Tutor was to give students a place where they can ask questions, receive explanations, practise Cambridge 9618-style pseudocode, and visualize program logic through flowcharts.

What it does

AI Pseudocode Tutor is a desktop application designed to support students learning Cambridge 9618 pseudocode. Users can:

💬 Ask pseudocode and programming questions through an AI-powered chat interface 📖 Receive explanations and guidance for their questions 📷 Upload images of questions for use within the application 🌳 Enter pseudocode and generate a visual flowchart representation 🧭 Navigate easily between the Home, Chat, and Flowchart sections

The flowchart feature recognises common structures such as START/END, INPUT, OUTPUT, IF/ELSE statements, and loops, helping students connect written pseudocode with visual program logic.

How we built it

built the project using Python and PySide6 to create the desktop graphical user interface.

The application uses multiple pages, including a Home page, Chat page, and Flowchart page, connected through a navigation system using QStackedWidget. For the AI functionality, I integrated my existing Claude Code setup into the project so the application could communicate with an AI assistant and provide explanations for students' questions. For the flowchart generator, I created logic that reads pseudocode line by line, identifies keywords and program structures, and displays the corresponding steps visually. I also added support for conditional branches so users can see YES and NO paths for simple IF/ELSE statements.

Challenges we ran into

One of the biggest challenges was that I was building a complete desktop application while learning many of the tools and technologies along the way. I encountered several Python errors, including indentation and import issues, and had to learn how different files and classes in a larger project communicate with each other.

Integrating AI was also challenging. My existing access to Claude was through Claude Code, rather than a traditional API key, so I had to explore how my Python application could communicate with that setup. I also encountered a Windows character-encoding issue when communicating through the command line, which required debugging before the connection worked correctly.

Another challenge was designing the flowchart generator. Pseudocode can contain many different structures, and converting text into a meaningful visual representation requires recognising keywords, conditions, branches, and loops. I focused on building a functional foundation that could be expanded further after the hackathon.

Accomplishments that we're proud of

Designed a clean user interface. Shifted the project from a basic solver into a true educational tool that teaches strategy.

What we learned

This project taught me much more than just Python programming. I learned how to structure a multi-file software project, build graphical interfaces with PySide6, connect different pages together, work with Git and GitHub, debug real-world errors, and integrate AI functionality into an application. Most importantly, I learned that building a project is an iterative process. Many features did not work perfectly on the first attempt, but debugging and improving them step by step helped turn an initial idea into a working application.

What's next for AI Pseudocode tutor

This project is only the beginning. I plan to continue developing AI Pseudocode Tutor after the hackathon for use within my school. Future improvements could include more advanced flowcharts, conversation history, improved image processing, exporting solutions, and additional tools specifically designed around the Cambridge 9618 syllabus.

My goal is to continue turning it into a genuinely useful learning tool for students who want to understand pseudocode rather than just memorise answers.

Built With

• git
• pyside6
• python
• python-dotenv
• venv
• visual-studio
• windows