Innovating EdTech with AI: Reflections from our June hackathon

As part of London Tech Week, we hosted our third AI Education Content Store hackathon. Designed to be fast-paced and collaborative, the event brought together some of the most creative minds in EdTech.

We built on the momentum of our January and March events and invited selected teams from our previous cohort to return with the opportunity to revisit, refine and reimagine their earlier ideas. 

Over a single afternoon, teams iterated on tools they’d prototyped in March – using the opportunity to integrate feedback and apply deeper pedagogical thinking. 

The challenge? To demonstrate innovation, effective use of the Content Store and strong pedagogical alignment (all within a few short hours). The three assessment and feedback use cases worked on by the teams were:

• Generating targeted formative assessments at a given point within a scheme of learning. Improving on a system that generates various examples of formative assessment for concepts covered within the KS3 Maths curriculum.

• Marking long-form, subjective assessments. Building on a tool which evaluates GCSE English Language written work against the necessary requirements.

• Providing targeted feedback in line with best practice. Developing on a platform able to provide targeted feedback on KS1/KS2 schoolwork while surfacing the best-practice sources referenced in its creation.

Each team presented their progress to a judging panel of sector experts on day one. On day two, video overviews of their demos were played to a broader audience of policymakers, technologists and the public on London Tech Week’s Tech Horizons stage. 

We're thrilled to share that Team 2 were named winners for their creative and thoughtful improvements which reflected the spirit of experimentation, collaboration and user-centred design that defines these events. In particular, Team 2’s GCSE English marking tool was praised for its use of the Content Store to deliver both measurable improvements in marking accuracy (a 32% increase when using Content Store data) and more nuanced, qualitative enhancements to the feedback generated for students.

Opening the doors to the public

One of our core goals with the third Content Store hackathon was to bring the public with us on the journey of developing the AI Education Content Store. For the first time, members of the general London Tech Week audience were able to see the outputs of our hackathon series live on stage, and gain insight into how making structured educational content openly accessible can underpin the development of AI tools which produce high quality outputs for users.

We were proud to showcase how the Department for Education (DfE) and the Department for Science, Innovation and Technology (DSIT) are helping to stimulate innovation in the EdTech space by enabling access to high-quality, structured educational data.

In the same session, our co-founder Angie Ma had the privilege of chairing a panel discussion on AI, innovation, and the future of education, joined by:

• Stephen Morgan MP, Parliamentary Under-Secretary of State for Early Education

• Professor Mutlu Cukurova, Professor of of Learning and Artificial Intelligence, UCL

• Sir Anthony Seldon, education historian and author

Their conversation ranged from the real-world applications of AI in schools, to the ethical guardrails needed to ensure equity, safety and pedagogical value. 

There was a shared understanding that innovation must take place in collaboration with educators, and a deep recognition of the significant role the public sector can play in enabling that innovation to thrive.

Outcomes we’re proud of

Our third hackathon marked another important step forward for the Content Store project. 

1. We broadened and deepened our engagement with the EdTech community; helping us build a growing network of advocates who see the value in open, structured educational content.

2. We demonstrated the viability of the Content Store in solving real problems in education (not just in theory, but in practice).

3. We opened up our work to a broader public audience: sparking new conversations about the role of AI in schools and the importance of transparent, government-backed data infrastructure.

What’s next?

We are taking the learnings from this hackathon and feeding them directly into the next phase of development. While continuing to work closely with DfE, DSIT and the wider EdTech community to explore the responsible and effective use of AI in classrooms. These events help us keep the needs of teachers and students front and centre as we iterate, test, and improve the tools that power educational innovation.

This third hackathon showed us that there’s real momentum and that each round of collaboration deepens our understanding of what EdTechs and educators need most from us. The use cases, prototypes and feedback received are already informing our design and development choices, ensuring we stay grounded in real-world needs as we refine the platform and extend its capabilities.