Applying prompts to GenAI/LLMs

A proof of concept argument for student-centred module leaders A tweet recently caught my attention  https://x.com/ihtesham2005/status/2041576806810370553?s=20 . It described an MIT student who had developed what he called “context stacking” — uploading lecture materials, readings and related papers into an AI tool before each class, then using carefully constructed prompts to build a mental model of the content before setting foot in the lecture hall. By the time he arrived, the professor wasn’t teaching him anything new. They were confirming, refining and occasionally surprising him. That surprise, he said, was the only thing he wrote down. This is not simply pre-reading with extra steps. Using generative AI as an external thinking partner, this student was identifying gaps in his own understanding before the lecture began — doing what good tutors have always done, asking not “what do you know?” but “where does your understanding break down?” This maps directly onto the highe...

If you supervise research students at undergraduate or postgraduate level, you are likely to be very familiar with the "blank stare"—that moment a student first confronts the sheer, overwhelming mountain of academic literature they are expected to read, synthesise, and critique. Information overload is a genuine academic pain point, often manifesting as a severe case of "blank page" syndrome. As academics, we know that starting is half the battle . This is where Generative AI, like ChatGPT-4o, shines not as a tool to write the review for the student, but as a structural scaffold. Much like using AI as a mirror to transform vague student ideas , we can use GenAI to help students map the thematic landscape of a topic before they dive into deep reading. It breaks the ice, organises chaos into a digestible format, and gives them a structured starting point. Here is a practical, step-by-step workflow you can share with your students to help them generate a foundational...

System Overview (produced using ChatGPT) There's a moment many of us have had: you're looking at a competitor's website, then back at your own, and you just know something's different — but you can't quite put your finger on what. Traditionally, getting a rigorous answer meant hiring a consultant, running expensive user research, or spending hours doing it manually. What if you could get sharp, structured, comparative analysis in under ten minutes — without writing a single line of code? That's exactly what this project set out to prove. It started with a specific problem: comparing course websites to understand how they stacked up against a competitor. The goal wasn't just a surface-level look — it was to understand how real people, with different needs and backgrounds, would actually experience each site. The solution turned out to be a structured ChatGPT workflow built entirely in standard chat, using nothing more than a sequence of carefully designed ...

The Problem: The "Generic App" and the "Time Sink" We’ve all been there: a student walks into a 1-to-1 with a vague desire to "do something with AI" or "build a fitness app." You spend 45 minutes trying to find a technical "hook" that justifies a Level 6 or Level 7 grade, only for the student to drift back into "CRUD app" territory by week three. The Philosophy: AI as a Mirror Instead of you doing the heavy lifting, this workflow uses AI as a Mirror . It reflects the student’s own skills and career goals back to them, but with the structural rigour of a virtual supervisory team. It’s not about the AI "giving" the idea; it’s about the AI forcing the student to defend and refine their own concepts until they hold water. The Framework: 3 Months of Rigour This prompt is specifically designed for intensive/conversion MSc or summer capstone projects . It assumes a tight 12-week implementation window. By forcing the AI to w...

There is no shortage of articles telling academics that Generative AI is going to transform education. It is, and it will continue to do so. However, many of these pieces are long on enthusiasm and short on detail. This is not one of those. What follows is a practical account of using ChatGPT to build a real teaching session. I’ll cover what I did, what worked, what failed, and how long it actually took. No hype—just the reality of how it saved me time and how it could possibly do the same for you. The Test Case My subject was a four-hour session on Pytest in Django , aimed at final-year BSc Software Engineering students. These students have a basic grasp of Django but possess solid overall coding skills. The session was split into a one-hour lecture and three hours of hands-on practical work in VS Code. The Strategy: Starting with the Prompt The key to getting useful output is being specific upfront. Rather than simply asking ChatGPT to "create a lesson on Pytest," I provide...

I've been learning Python data visualisation, working through Murat Durmus's Hands-On Introduction to Essential Python Libraries and Frameworks alongside the official Dash tutorial . The resulting code was functional — a basic bar chart comparing data for San Francisco and Montréal — but it looked like exactly what it was: a beginner's first attempt. Plain white background, default colours, numbered axes, and a title that just said "Data Viz." So I decided to run an experiment. Could Claude AI turn a scrappy 20-line script into something genuinely worth showing people? Before running the prompt The First Prompt I pasted the code into Claude.ai with a simple instruction: "Rewrite this following code to be graphically more interesting." The result was striking. Claude switched to a dark "neon terminal" aesthetic — deep navy background, electric teal and magenta accents, and a stylish monospaced font. The bars got proper labels, the axe...