// Academic Project · Year 2 · COMP2912 · Software Design Brief
AI-Powered Mental Health App Design
Designed an AI-assisted mental health app for university students, including how their data is kept private.
A software design brief for a £100,000 mental health app for university students around the world. The design centers on a ChatGPT-4 Turbo companion and covers the stakeholders, the system architecture, links into university systems, crisis intervention and privacy controls.
Business Context & Requirements
COMP2912 Software Engineering Principles
The scenario: a mental health charity has £100,000 to build an app and runs a tender, with a £10,000 prize for the best design. Each entry is a 6-page technical specification. The app had to support student mental health at universities worldwide, with personalized AI interventions, healthy lifestyle prompts and a way to handle crises.
Business Requirements
- • £100,000 development budget
- • Rollout to universities worldwide
- • Competitive tender (£10,000 prize)
- • 6-page design specification
Core Focus Areas
- • Mental health support for university students
- • Personalized interventions driven by AI
- • Better sleep, exercise and nutrition
- • Crisis intervention and links to healthcare services
Stakeholder Analysis & Requirements Engineering
Primary Users
- • They prefer a personal approach
- • They want support they can trust, without judgment
- • Academic and money worries
- • Social and relationship problems
Institutional Stakeholders
- • Better student performance
- • Professional endorsement
- • Integration with their systems
- • Funding that lasts
Regulatory & Ethics
- • Data Protection Act
- • GDPR
- • AI legislation
- • Privacy and security standards
Use Case Analysis & Value Propositions
Primary Use Cases
- • Talk to AI: A companion to talk to, 24/7
- • Crisis Intervention: Puts the user in touch with help automatically
- • Tips & Tricks: Wellbeing notifications before problems build up
- • Sleep Monitoring: Helps keep a healthy sleep cycle
System Integration
- • University Gateway: Students sign in through their university
- • Healthcare Services: BetterHelp and the NHS
- • Timetable Sync: Knows the student's academic schedule
- • Minimalist Mode: Cuts distractions so students can focus
AI Architecture
ChatGPT-4 Turbo Integration
- • Learns a profile of each user
- • Aware of cultural and demographic differences
- • Detects a crisis and responds to it
- • Suggests resources that fit the conversation
Smart Interventions
- • Coping with academic pressure
- • Guidance on money worries
- • Support with relationships
- • Suicide prevention protocols
System Architecture & Implementation
API Integration: ChatGPT-4 Turbo + University Gateways
Data Storage: Local encryption + secure transmission
Healthcare APIs: BetterHelp, NHS Crisis Line, 999
Security: 2FA + GDPR compliance + data anonymization
Performance
Fast native languages, small data packets and efficient API calls
Security
End-to-end encryption, data stored on the device, two-factor authentication
Scalability
Worldwide rollout, links into university systems, support for multiple languages
Software Engineering
Methodology
- Agile Development: Delivered in increments
- User Involvement: Regular feedback from stakeholders
- Iterative Design: A three-phase investigation
- Quality Assurance: Thorough testing and validation
- Documentation: Detailed technical specifications
UML & Design
- Use Case Diagrams: Every actor in the system
- Implementation Diagrams: Deployment and component views
- User Stories: Step-by-step interaction scenarios
- Prototyping: UI mockups and workflow diagrams
- Architecture Design: A multi-tier system specification
Engineering Practices
- Requirements Engineering: A structured elicitation process
- Risk Assessment: Technical and business risks
- Cost Estimation: Planning within the budget
- Timeline Management: Realistic milestones
- Industry Standards: Standard professional practice
Ethics & Legal Compliance Framework
Ethical Considerations
- • Discrimination Prevention: Protocols for cultural sensitivity
- • Addiction Management: Monitoring how much the app is used
- • Privacy Protection: User consent, and users own their data
- • Transparency: Users always know they are talking to an AI
Legal Compliance
- • Data Protection Act: UK privacy law
- • GDPR: European data protection standards
- • AI Legislation: New and upcoming regulation
- • Healthcare Standards: Rules for handling medical data
Design Brief & Takeaways
Documentation
- 6-Page Design Brief: Written as if for a real client
- Requirements: Functional and non-functional
- Technical Architecture: A full implementation roadmap
- UML Diagrams: Use case and deployment models
- User Stories: Step-by-step interaction scenarios
- Business Analysis: Cost, timeline and feasibility
Academic Achievement
- Tender-Style Brief: Written to the standard of a real tender
- Multi-Stakeholder Analysis: Requirements gathered from every group
- AI Integration: Built around ChatGPT-4 Turbo
- Global Scale: Planned for rollout to universities worldwide
- Ethics: Privacy and user safety built into the design
- Business Awareness: Worked within a real budget and timeline
Download Complete Design Brief
Read the full 6-page design brief I submitted for the COMP2912 Software Engineering Principles coursework, alongside the assessment brief it answers.
Covers requirements, UML models, architecture and the ethics of the design
Key Insights
- • Designing software within real-world constraints
- • Planning how to use AI safely in a healthcare app
- • Stakeholder analysis and requirements engineering
- • Making privacy and security part of the design from the start
- • Planning a system to run at a worldwide scale