Skip to main content

Prompt Engineering, Context Engineering and Agentic AI in Higher Education: A Short Briefing

AI in education


Prompt engineering, context engineering and agentic AI are often used interchangeably, but the literature treats them as distinct. Prompt engineering concerns crafting effective single instructions to a model (Glean, 2026). Context engineering is the broader discipline of designing and managing the entire informational environment around a model—memory, retrieval, tool outputs and conversation state—rather than a one-off instruction (Abstracta, 2026); prompt engineering operates within the context window, while context engineering determines what fills it (arXiv:2606.12422, 2026). Agentic AI describes systems that plan and execute multi-step tasks with delegated autonomy, raising organisational questions of accountability rather than purely technical ones (MIT Sloan, 2026; Palo Alto Networks, 2026).

There is genuine debate about prompt engineering's durability. IEEE Spectrum (2025) reported research suggesting prompting is increasingly performed by models themselves, and standalone "Prompt Engineer" job postings have fallen sharply since 2024 (Future Factors, 2026). Yet roles requiring prompting skill are reported to have grown threefold over the same period, with the more persuasive reading being that the skill has become embedded foundational literacy rather than disappearing (Unrot, 2026). Meanwhile, industry data shows most leaders now consider prompting alone insufficient, with the discipline shifting toward context engineering to support agentic systems at scale (Glean, 2026).

Agentic AI is not purely a CS problem. Because agents act under delegated authority, governance structures are widely argued to need cross-functional representation spanning technology, law, ethics and business operations, since no single function has the expertise to govern autonomous systems alone (Agility at Scale, 2026). Delegation does not remove accountability: organisations remain responsible for the permissions and scope they grant (Palo Alto Networks, 2026). Notably, MIT Sloan (2026) found the dominant challenge in agentic AI deployment was not prompting or fine-tuning but data engineering, stakeholder alignment and governance—supporting the case that engineering and organisational competencies cannot be taught in isolation from one another.

For CS students, this suggests all three areas belong in the curriculum, but with different depth: prompt engineering as brief foundational literacy; context engineering taught in depth as a systems discipline; and agentic AI taught with equal weight on technical build and governance/accountability, consistent with the QAA Computing benchmark's expectation that design decisions be justified through argument, not treated as purely technical (QAA, 2022), and the BCS Code of Conduct's expectations of professional accountability (BCS, n.d.). However, benchmark statements deliberately avoid specifying a syllabus, so ethics teaching varies in depth between providers unless deliberately spiralled across levels.

For non-CS students, evidence from discipline-agnostic AI literacy courses shows meaningful gains in knowledge and ethical reasoning are achievable without technical prerequisites (Gogovi, 2026; arXiv:2604.09634, 2026), and a twelve-competency literacy framework has been proposed to scaffold this across all subject areas (arXiv:2412.12107, 2024). The reasonable target is functional prompt literacy, conceptual understanding of why context affects reliability, and governance awareness of agentic delegation relevant to their future roles overseeing rather than building such systems.

Bridging the two populations is best attempted through structured interdisciplinary project work, which measurably improves collaboration skills (MDPI, 2024) and multidisciplinary team functioning (arXiv:1410.6935, 2014). However, unstructured collaboration risks role-siloing and free-riding, with business students defaulting to management roles and technical students remaining isolated (Journal of Education for Business, 2023)—meaning deliberate assessment and role design is essential, not optional.

Reference List

Abstracta (2026) Context Engineering vs Prompt Engineering. Available at: https://abstracta.us/blog/ai/context-engineering-vs-prompt-engineering/ (Accessed: 7 July 2026).

Agility at Scale (2026) Agentic AI Governance: Securing Autonomous AI Agents in Enterprise. Available at: https://agility-at-scale.com/ai/governance/agentic-ai-governance/ (Accessed: 7 July 2026).

BCS, The Chartered Institute for IT (n.d.) BCS Code of Conduct. Available at: https://www.bcs.org/media/2211/bcs-code-of-conduct.pdf (Accessed: 7 July 2026).

Future Factors (2026) Is Prompt Engineering Dead? What Replaced It. Available at: https://futurefactors.ai/prompt-engineering-dead-what-replaced-it-2026/ (Accessed: 7 July 2026).

Glean (2026) Context engineering vs prompt engineering: key differences explained. Available at: https://www.glean.com/perspectives/context-engineering-vs-prompt-engineering-key-differences-explained (Accessed: 7 July 2026).

Gogovi, G.K. (2026) 'A Discipline-Agnostic AI Literacy Course for Academic Research', arXiv preprint arXiv:2604.27225. Available at: https://arxiv.org/pdf/2604.27225 (Accessed: 7 July 2026).

IEEE Spectrum (2025) AI Prompt Engineering Is Dead. Available at: https://spectrum.ieee.org/prompt-engineering-is-dead (Accessed: 7 July 2026).

Journal of Education for Business (2023) 'Learning by facilitating: A project-based interdisciplinary approach'. Available at: https://www.tandfonline.com/doi/full/10.1080/08832323.2023.2196049 (Accessed: 7 July 2026).

MDPI (2024) 'Transforming Learning Orientations Through STEM Interdisciplinary Project-Based Learning', Education Sciences, 14(11), p.1154. Available at: https://www.mdpi.com/2227-7102/14/11/1154 (Accessed: 7 July 2026).

MIT Sloan (2026) Agentic AI, explained. Available at: https://mitsloan.mit.edu/ideas-made-to-matter/agentic-ai-explained (Accessed: 7 July 2026).

Palo Alto Networks (2026) A Complete Guide to Agentic AI Governance. Available at: https://www.paloaltonetworks.com/cyberpedia/what-is-agentic-ai-governance (Accessed: 7 July 2026).

Quality Assurance Agency for Higher Education (QAA) (2022) Subject Benchmark Statement: Computing (including Master's). Available at: https://www.qaa.ac.uk/the-quality-code/subject-benchmark-statements/computing (Accessed: 7 July 2026).

Unrot (2026) Prompt Engineering: The Most In-Demand AI Skill of 2026. Available at: https://unrot.co/blogs/prompt-engineering-2026 (Accessed: 7 July 2026).

arXiv:1410.6935 (2014) 'Project-based Learning within a Large-Scale Interdisciplinary Research Effort'. Available at: https://arxiv.org/pdf/1410.6935 (Accessed: 7 July 2026).

arXiv:2412.12107 (2024) 'Generative AI Literacy: Twelve Defining Competencies'. Available at: https://arxiv.org/pdf/2412.12107 (Accessed: 7 July 2026).

arXiv:2604.09634 (2026) 'From Understanding to Creation: A Prerequisite-Free AI Literacy Course with Technical Depth Across Majors'. Available at: https://arxiv.org/pdf/2604.09634 (Accessed: 7 July 2026).

arXiv:2606.12422 (2026) 'Creating and Evaluating K-12 GenAI Assessment Graders Through Context Engineering'. Available at: https://arxiv.org/pdf/2606.12422 (Accessed: 7 July 2026).


All opinions in this blog are the Author's and should not in any way be seen as reflecting the views of any organisation the Author has any association with. Twitter @scottturneruon

Comments

Post a Comment

Popular posts from this blog

AI as a Mirror: Transforming Vague Student Ideas into a More Rigorous Project Agreement

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...
Post a Comment
Read more

Getting multiple viewpoints with ChatGPT

Well sort of! There are approaches where we can get the generative AI to look at a problem from multiple perspectives (or personas) and bring the ideas generated, ideally informed by the others. to a final plan. One of the main strategy is called Tree of Thoughts (see here for more detail  https://www.forbes.com/sites/lanceeliot/2023/09/08/prompt-engineering-embraces-tree-of-thoughts-as-latest-new-technique-to-solve-generative-ai-toughest-problems/?sh=5ce79bdb2c8b ). The central idea is get a number of expert opinions, allow potential cross-fertilization of ideas, come up with actions or plans. Let see this action.  Scenario: Find out about the UK Government's plans on Disability support and then use Tree of Thoughts to produce some ideas for a company making disability equipment based on their website. Google's Gemini will be used. Stage 1 "UK Governments plans on Disability support ": Prompt:  Read, convert to plain text and consolidate information from the followi...
Post a Comment
Read more

GenAI Productivity: Ideas to project proposal 1

One of the ways I use Generative AI with students is to take basic ideas for projects, usually a title, and get these tools to greater ideas and start of a project proposal. This is with all the usual caveats  Check the references (if any); It is going to be basic, so extend it. In this example I am going to use Co-pilot but the ChatGPT, etc can be used, employing a few basic prompt engineering basics: personas (who is the target audience?) and Templates (how do I want it to look?) to start this process. Example:  Project ideas for MSc Data Intelligence students (persona)  on a particular topic. The reply will include subheadings and relevant (hopefully) content for  TITLE, INTRODUCTION, PROBLEM STATEMENT. The prompt: " Taking the topic "Leveraging open-source tools to measure and present academics publications automatically from public domain data.". Give five innovative projects for a Master's level student dissertation in Data Intelligence. Each project example wi...
Post a Comment
Read more