01
Problem-led
We begin with the decision, workflow or user need—not with a predetermined tool. Success criteria and operating constraints are made visible before development begins.
Selected project portfolio · China
Applied AI, robotics and student innovation—from a defined challenge to a working demonstration.
This portfolio presents selected project types delivered in China across enterprise AI enablement, quadruped robot secondary development and team-based student app development. To respect confidentiality, the cases are described at programme level: they explain the challenge, technical work, collaboration model and outputs without naming clients or claiming unverified commercial results.
China project experience
AI enablement
Robot development
Student product teams
01 / Portfolio overview
Projects designed around real users, real constraints and demonstrable outcomes
CIRC projects bring together problem discovery, technical development, team capability and responsible implementation. The format changes by audience, but every engagement is structured to move beyond a one-off demonstration and leave behind reusable knowledge, evidence and practical next steps.
02
Prototype-driven
Teams make ideas testable early through workflow mock-ups, software prototypes, robot behaviours or minimum viable products.
03
Evidence-informed
User feedback, technical tests, design records and review criteria are used to distinguish a promising concept from a deployable solution.
04
Capability-building
Documentation, mentoring, demonstrations and handover activities help participants understand, maintain and extend what has been created.
02 / China case · Enterprise AI enablement
Turning broad AI ambition into focused business workflows
In this project type, a Chinese enterprise explores where generative AI, knowledge retrieval or intelligent automation can create practical value. The work is framed as an enablement programme: understand the organisation, select bounded use cases, build prototypes and prepare teams to evaluate them responsibly.
Challenge
From interest to a credible use case
Leaders and teams may recognise the importance of AI but need help identifying tasks with suitable data, acceptable risk and a clear human owner.
Discovery
Workflow and opportunity mapping
Interviews and working sessions map repetitive tasks, information bottlenecks, customer or staff journeys, data readiness and the points where human judgement must remain central.
Development
Prototype intelligent assistance
Depending on the use case, the team can prototype internal knowledge search, document summarisation, structured content generation, classification, workflow support or decision-preparation tools.
Governance
Design human review into the system
The project defines permissions, source traceability, review checkpoints, privacy boundaries, failure handling and the conditions under which an output must not be used.
Enterprise project outputs
What the organisation receives—and what the pilot is designed to prove
01
Prioritised opportunity map
A structured view of candidate workflows assessed against value, feasibility, data availability, user adoption and operational risk.
02
Working proof of concept
A bounded prototype or demonstrator that allows real users to test assumptions before a larger technology commitment is made.
03
Adoption and governance plan
Recommended ownership, review processes, training needs, risk controls and pilot measures for the next stage of implementation.
04
Team learning
Practical workshops help staff understand model limitations, prompt and evaluation methods, data responsibilities and appropriate human oversight.
03 / China case · Quadruped robot secondary development
Extending a robot dog from a hardware platform into a task-specific system
This project type focuses on secondary development of a commercial quadruped robot. Rather than treating the robot as a finished product, the project integrates software, sensors, interfaces and operating procedures around a defined education, inspection, demonstration or research scenario.
Scenario
Define the mission and environment
The team specifies the route, terrain, payload, interaction, connectivity, operating distance, safety boundary and success conditions before choosing features.
Integration
Work with SDKs, APIs and control layers
Developers connect supported control interfaces, telemetry and task commands, then create modular behaviours that can be tested independently.
Perception
Connect sensing to action
Where appropriate, camera or other sensor data can support object detection, environmental observation, marker recognition or operator awareness, with explicit limits on autonomy.
Interface
Make the system understandable to operators
A control dashboard can present robot status, live data, mission steps, alerts and manual override options for supervised operation.
Robot development and validation
A staged engineering process for reliability and safety
01
Architecture and simulation
Break the system into control, sensing, communication and application modules; test logic in simulation or bench conditions before full movement trials.
02
Controlled field testing
Use repeatable test routes and checklists to examine stability, latency, connection loss, obstacle response, battery constraints and operator intervention.
03
Safety and human control
Define exclusion zones, emergency-stop procedures, supervision, speed limits, equipment checks and fallback behaviours appropriate to the operating context.
04
Technical handover
Provide system diagrams, configuration records, source or module documentation, operating guidance, test evidence and recommendations for the next development cycle.
04 / China case · Student team-based app development
Students work as product teams to research, design, build and present an app
In this project-based learning model, students respond to a meaningful brief and develop a mobile or web application in groups. The emphasis is not only on coding: learners practise user research, product decisions, interface design, data thinking, testing, communication and accountable teamwork.
Discover
Investigate a real user problem
Teams select or receive a brief—such as campus services, learning support, sustainability, wellbeing or accessibility—and gather evidence about users and context.
Design
Convert insight into a product concept
Students define user stories, prioritise features, map the experience and create wireframes or clickable prototypes before building.
Build
Develop a minimum viable product
Using age-appropriate code, low-code tools or APIs, teams implement the essential journey, manage data carefully and integrate AI only where it has a justified role.
Validate
Test, improve and communicate
Peers and intended users test the app. Teams log issues, refine the design, explain limitations and present a live demonstration supported by evidence.
Student team structure
Clear roles create both individual accountability and collective ownership
01
Product and research
Frames the problem, organises user evidence, maintains the feature priorities and checks that decisions serve the agreed need.
02
UX and interface design
Develops information architecture, screen flows, visual consistency and accessibility, then responds to usability testing.
03
Development and data
Builds front-end and back-end functions, handles data structures or AI integrations and documents technical choices.
04
Quality and communication
Coordinates test cases, records defects, checks evidence and leads the final demonstration, pitch and team reflection.
Student learning evidence
Assessment captures the process as well as the final app
01
Product evidence
Problem statement, user research, user stories, prototype, working features, test record and final demonstration.
02
Technical evidence
Architecture or data model, code or build logic, responsible-AI decisions, known limitations and technical documentation.
03
Team evidence
Role records, sprint reviews, task board, peer feedback, decision log and individual reflection on contribution and learning.
04
Presentation evidence
A concise pitch explains the need, solution, user feedback, value, limitations and the next version the team would build.
05 / Common delivery framework
A repeatable route from opportunity to a tested outcome
01
Discover
Clarify stakeholders, users, operating context, constraints, available data and measurable success criteria.
02
Co-design
Select a bounded project, define roles and architecture, agree responsible-use rules and plan how evidence will be collected.
03
Prototype and validate
Build in short cycles, test with users or in controlled environments, record failures and improve the most important assumptions.
04
Demonstrate and hand over
Present the result, document decisions and limitations, transfer capability and define the next development or deployment stage.
06 / What this portfolio demonstrates
One centre working across organisational innovation, physical AI and youth capability
01
Enterprise application
The ability to translate AI into bounded workflows, prototypes, governance decisions and staff capability.
02
Robotics engineering
The ability to integrate hardware and software around a scenario, while maintaining staged testing and meaningful human control.
03
Project-based education
The ability to structure authentic product work so that young people develop technical, creative and collaborative capability.
04
International potential
The project formats can support future UK–China exchange, comparative research, joint challenges and cross-border student or professional teams.
Build the next programme
Ready to turn AI and robotics into purposeful student work?
We can shape a class project, school pilot, international team or multi-school pathway around your learners.