Cambridge, UNICEF and GPE · 2025 to 2026
Turning global frameworks into national decisions.
A five-month leadership programme for more than 40 education leaders from 17 countries, built around practical system analysis, cross-country learning and national action planning.1
My role: AI Lead for the programme at Cambridge University Press & Assessment. I helped shape the learning architecture, led Cambridge’s technical contribution and delivery coordination, supported content and applied-activity design, and facilitated live learning with the wider programme team.
The challenge
Digital education is a system problem, not a technology purchase.
Education leaders were making connected decisions about policy, teacher capability, digital content, data and AI, connectivity, procurement and long-term financing. These decisions were often being made under tight budgets, uneven infrastructure and rapid technological change.
DELP was co-designed by Cambridge, UNICEF and the Global Partnership for Education to help leaders analyse and strengthen national digital-education plans, identify gaps and opportunities, and turn shared frameworks into action that made sense in their own contexts.1
Core design question
How could a global programme give leaders useful common frameworks without flattening the policy, infrastructure, institutional and classroom realities of 17 different country contexts?

My contribution
Learning architecture, delivery and practical application
The work combined programme leadership with detailed learning design. The aim was not simply to transmit knowledge, but to help participants use evidence and frameworks to make better system decisions.
Shape the learning journey
Helped translate the Six Pillars Framework into a seven-module progression that connected leadership, capacity, content, data and AI, infrastructure and financing.
Keep application central
Supported module activities and a cumulative capstone so that participants tested global evidence against their national priorities rather than treating the programme as a sequence of lectures.
Coordinate specialist delivery
Worked across Cambridge, UNICEF and GPE colleagues, programme specialists and expert speakers to connect technical content, facilitation and participant needs.
Use AI with judgement in the loop
Integrated guided AI-supported research and synthesis where useful, while keeping contextual analysis, verification and professional judgement with participants.
Programme architecture
Seven connected modules
The blended programme combined online learning, question-and-answer sessions, global case studies and live lectures, with added emphasis on AI.1

Programme evidence
Early application in country work
The available evidence shows participants applying the programme to live policy and implementation questions. It is not yet a longitudinal measure of system or learner impact.
These examples are reported in Cambridge’s public programme story and are attributed to the participants and institutions involved.2
What I took from the work
Three design lessons
- Frameworks matter when they structure real decisions. Their value comes from helping teams compare options, expose assumptions and identify the next practical move.
- Peer exchange is part of the learning architecture. Cross-country discussion helped leaders test whether an approach travelled, what needed adaptation and where common problems had different causes.
- Sustainability must shape the brief early. Financing, maintenance, governance and institutional capacity cannot be left until after a technology choice has been made.
Attribution: DELP was a collective programme co-designed by Cambridge, UNICEF and GPE. This case study focuses on my contribution as Cambridge’s AI Lead for the programme and does not imply sole authorship of the curriculum, delivery or outcomes. Programme images are used from the public Cambridge story or from the working programme asset set.
Sources and further reading
Explore more work
See other projects spanning AI-assisted publishing, professional learning and education-system transformation.