Prompt and Context Engineering
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.
The workshop began with foundations covering AI’s strengths in pattern recognition, repetitive tasks, and scaling versus its limitations in cultural nuance, emotional intelligence, and contextual judgement, before introducing the Human-AI Partnership Framework (Define → Draft → Refine → Validate → Deliver) and prompt engineering using the CRAFT method (Context, Role, Action, Format, Tone). Participants learned to identify high-impact use cases through a 4-question test evaluating whether tasks are time-intensive, repeatable, scalable, and expertise-based.
The second phase moved from prompt to production, teaching the 7-Step AI Workflow Framework for building reusable systems, with particular emphasis on context engineering using knowledge files specific to Kuwait MOE standards, local cultural references, and curriculum requirements. Teachers learned to design structured outputs for lesson plans, assessments, and feedback that maintain consistency, while building quality checkpoints combining automated AI validation with human review for cultural sensitivity and practical feasibility, culminating in a live demonstration of a Formative Assessment Generator with Kuwait-specific context files.
During the hands-on activity, teachers worked through a guided worksheet to select one repetitive task from their workload, score it using the use case framework, write a complete CRAFT prompt with Kuwait-specific context, and share and refine with peers. Participants demonstrated high engagement and technical readiness, with the cohort already using AI tools and expressing strong interest in advancing to building personalized AI systems tailored to Kuwait’s educational context, setting up AI agents for automated tasks like student feedback and differentiation strategies, and moving beyond prompts to sustainable, production-ready workflows. This workshop laid the groundwork for educators to transition from experimental AI use to systematic implementation that respects local context while reducing repetitive workload.