Prompt engineering
Designing prompts that produce useful, traceable summaries — and testing what happens when they fail.
A proposed Wundr Space programme
Build useful technology. Develop lasting capability.
Salterns Signal Lab brings students, staff, families and technology professionals together to address a real challenge within their school community — building practical experience in AI, software engineering, data protection and cyber security along the way.
This is an early-stage programme being developed with schools and communities.
The central idea
Not another platform for schools to adapt to.
Schools do not need another disconnected system imposed upon them. The Signal Lab starts with the school's people, context and existing ways of working. Together, participants investigate a genuine need, design an appropriate response and decide whether technology should play a role.
People adapting to softwareTechnology shaped around people, place and purpose
“The application is an outcome of the learning — not the sole purpose of it.”
- Tangible outcomes co-designed with the school community
- Hands-on capability growth for students and staff
- Better questions asked before technology choices are made
- Reusable knowledge, artefacts and evidence
- Technology shaped around the school, not the other way around
The first challenge
A real challenge: connecting schools and communities.
Field-journal enquiry
How might we help every family remain connected to school life, while enabling schools to listen responsibly and respond to emerging community needs?
Students would help investigate whether this is desirable, appropriate and achievable. The programme does not assume the original solution is automatically correct.
Observed tensions and early routes
01School information is distributed across websites, newsletters, applications and social media.
02Families who do not use social media may become disconnected.
03Important posts can be hidden by platform algorithms.
04Schools may not recognise repeated questions or misunderstandings early.
05Positive community stories and advocacy can quickly disappear.
06Listening can become intrusive unless privacy, proportionality and human judgement are designed in from the start.
Bring together school-controlled communications.Create relevant parent and carer digests.Help authorised staff recognise recurring themes and questions.Surface positive community contributions.Support earlier clarification and engagement.Keep interpretation and decision-making firmly with people.
How the programme works
A journey of enquiry, not a conveyor belt.
The programme moves from listening to questioning, designing and building — then returns evidence the school can act on.
- A01
Discover
Map how communication currently works and learn from students, staff and families.
Evidence: User interviews · Communication-channel mapping · Stakeholder analysis - B02
Question
Examine the ethical, social and practical questions before choosing technology.
Evidence: Should this information be collected? · What would families reasonably expect? · Who owns the data? - C03
Design
Create user journeys, prototypes, data flows and responsible-technology principles.
Evidence: Parent-digest prototype · Leadership insight concept · Threat model - D04
Build
Develop selected components using synthetic or appropriately controlled data.
Evidence: News aggregation prototype · Personalised digest · Accessible user interface - E05
Assure
Test security, privacy, accessibility, AI behaviour and potential misuse.
Evidence: Code review · Automated testing · Accessibility evaluation - F06
Demonstrate & decide
Present the evidence to the school community and decide what, if anything, should progress.
Evidence: Student demonstration · Evaluation report · Lessons learned
What students learn
An interconnected learning map around one authentic challenge.
These disciplines are not separate modules. They meet wherever students need to make a responsible product decision.
School-community communication challenge
Responsible AI
Exploring bias, hallucination, transparency, proportionality and human oversight.
Product discovery
Learning from real users before deciding what to build.
Software engineering
Working with branches, pull requests, automated tests, peer review and documented decisions.
Data protection
Understanding collection, consent, lawful use, retention, deletion and individual rights.
Cyber security
Applying threat modelling, access control, dependency management and secure development.
Data ownership
Examining who owns source content, derived insights, software and community information.
UX & accessibility
Designing for different abilities, circumstances and levels of digital confidence.
Digital citizenship
Understanding platforms, algorithms, misinformation, participation and responsibility.
Communication & governance
Explaining technical choices and presenting evidence to staff, parents, leaders and trustees.
- Prompt engineering: Designing prompts that produce useful, traceable summaries — and testing what happens when they fail.
- Responsible AI: Exploring bias, hallucination, transparency, proportionality and human oversight.
- Product discovery: Learning from real users before deciding what to build.
- Software engineering: Working with branches, pull requests, automated tests, peer review and documented decisions.
- Data protection: Understanding collection, consent, lawful use, retention, deletion and individual rights.
- Cyber security: Applying threat modelling, access control, dependency management and secure development.
- Data ownership: Examining who owns source content, derived insights, software and community information.
- UX & accessibility: Designing for different abilities, circumstances and levels of digital confidence.
- Digital citizenship: Understanding platforms, algorithms, misinformation, participation and responsibility.
- Communication & governance: Explaining technical choices and presenting evidence to staff, parents, leaders and trustees.
A real engineering environment
Students contribute. Professionals remain accountable.
The codebase is managed through a professionally governed GitHub organisation. Students work like engineers — with safeguards appropriate to their age and the domain.
- 01
Student feature branch
opened - 02
Pull request
opened - 03
Automated checks
reviewed - 04
Peer & mentor review
reviewed - 05
Wundr Space approval
reviewed - 06
Protected main branch
controlled - 07
Controlled test environment
controlled - 08
Authorised production decision
controlled
learning permissions
Students can
Create branches · Build features · Write documentation · Add tests · Open pull requests · Review one another's work · Respond to professional feedback · See how automated controls evaluate their work
protected controls
Students cannot
Push directly to the protected main branch · Access production credentials · Work with identifiable parent or safeguarding data · Disable security checks · Approve their own production changes · Deploy independently to a live environment
Safety and data boundaries
Authentic work does not require unrestricted access.
Learning side
Student learning environment
- Synthetic or anonymised datasets
- Prototype integrations
- Simulated community discussions
- Experimental AI analysis
- Learning-focused security exercises
- Isolated development environments
Operational side
Controlled operational environment
- Authorised school data
- Protected credentials
- Approved integrations
- Human-reviewed outputs
- Professional security controls
- Defined safeguarding and incident processes
- Students would not initially access identifiable parent discussion.
- Private parent groups would not be entered or scraped.
- Safeguarding data would remain outside the learning environment.
- AI would identify themes, not classify individual parents.
- Production decisions would remain with authorised adults.
- Data protection and safeguarding leads would be involved from the outset.
Responsible technology principles
Responsibility is part of the build.
“We do not teach students simply how to use AI. We help them understand when, why and whether it should be used.”
- 01Human first
People — students, staff and families — set the direction. Technology follows.
- 02Context over software
The school's culture and constraints shape the solution, not the reverse.
- 03Privacy & security by design
Protection is designed in from day one, not bolted on later.
- 04Transparency
How the technology works, and why, is explainable to non-technical audiences.
- 05Accountability
Responsibility for decisions and outputs remains with named people.
- 06Fairness
We look actively for who benefits, who is missed and who could be harmed.
- 07Proportionality
We use the least intrusive option that meets the need.
- 08Evidence before automation
We prove value with people before scaling with software.
- 09Accessibility & inclusion
Different abilities, languages and circumstances are considered from the start.
- 10Human review
AI output supports human judgement; it does not replace it.
- 11Data minimisation
We collect only what is necessary, and keep it only as long as needed.
- 12Continual learning
We revisit choices as evidence, technology and community needs evolve.
Value for each participant
One shared endeavour, different forms of value.
Students
Build something meaningful, develop current technical and professional skills, and create evidence of real project experience.
Schools
Explore a genuine operational or community challenge while developing internal digital capability.
Teachers
Connect curriculum knowledge to a professionally facilitated, multidisciplinary project.
Parents & carers
Help shape technology intended to make school communication more inclusive and responsive.
Leaders & trustees
See responsible innovation, student development and community engagement brought together within a governed programme.
Beyond the first challenge
A model that can address other school priorities.
School communication is our starting point, not the whole story. The same method applies wherever a school has a genuine challenge worth exploring carefully. Schools are welcome to bring their own.
- Primary-to-secondary transition
- Student voice
- Enrichment discovery
- Accessibility and inclusion
- Understanding school policies
- Reducing food or energy waste
- Alumni and community engagement
- Preserving institutional knowledge
- Supporting local partnerships
Pilot format
Explore a Signal Lab pilot.
These are indicative programme formats. The actual shape of a pilot would be designed with the school. We do not publish invented pricing.
Pilot enquiry
Bring a real challenge to your students.
We're inviting schools and multi-academy trusts to help shape the first Signal Lab pilots. Tell us about your school, your students and the challenge you would like to explore.
What happens next
We review what you share and, if the fit looks promising, get back in touch to arrange a conversation. Submitting this form does not confirm a pilot place.
The current service name is provisional and may change as the programme is shaped with schools.
Community interest
Register community interest
For parents, students, teachers or community partners who would like to help shape the programme.
Frequently asked questions
Honest answers to the questions we're asked.
Have a question that isn't answered here? Use the community form or write to us — we're actively shaping this programme with schools and would rather answer questions directly.