Encoding professional expertise into AI.
I write about four connected themes: AI readiness in education, expertise encoding and knowledge work, building AI products in real contexts, and Global South perspectives on AI implementation.
Start here
Your School Isn’t Behind on AI. You Just Don’t Have a Map Yet.
The question is not which AI tools to buy
What 28 AI Readiness Assessments Reveal About Schools in Ghana
NiallOS: a personal AI operating system for knowledge work
Published on Medium for long-form articles and Substack for newsletter essays. For shorter takes and daily conversation, find me on LinkedIn.
I was in Kuwait last week running an intensive three-day workshop with Kuwait Foundation for the Advancement of Sciences (KFAS) and Cambridge University Press & Assessment to equip Kuwaiti educators with practical frameworks for implementing AI in their teaching contexts. Teachers were separated into three tracks (Higher Education, Secondary, and Primary) and rotated through hands-on sessions focused on building sustainable, context-aware AI workflows.
I was invited to Kuwait University during Science Month to deliver a workshop on AI prompting and context engineering for Life Sciences faculty. The session explored practical applications of generative AI in higher education, with participants developing personalised AI agents to streamline administrative workflows, assessment design, and student feedback processes.
If you’ve been using AI tools like ChatGPT or Claude, you’ve probably experienced this frustration: every time you start a new conversation, you have to re-explain who you are, what you need, and the context for your work. It’s like introducing yourself to the same colleague every single day.
This is the fundamental limitation of prompt engineering – and it’s why the field is rapidly evolving toward something more powerful: context engineering.
Prompt engineering has been the dominant approach to working with AI: craft the perfect instructions, get better results. But here’s the challenge:
Every conversation starts from zero. You must re-specify your context, preferences, constraints, and requirements with every single prompt.
If you’re a teacher in South Africa working with CAPS curriculum, this might look like:
Research shows that 80% of prompt engineering effort goes into re-establishing basic context, not describing the actual task. This is inefficient, inconsistent, and doesn’t scale.