XRisis Project Commences Requirements Gathering with Action Contre la Faim

User-Centric Design Approach

The XRisis project has commenced comprehensive requirements gathering employing user-centric design methodologies that prioritise deep understanding of humanitarian operational contexts before proposing technical solutions. The Nuwa team is participating directly in Action Contre la Faim's Cross Knowledge e-learning modules covering emergency management concepts, organisational procedures, and coordination mechanisms, building familiarity with terminology, pedagogical approaches, and institutional workflows that inform effective platform design. Team members are observing four three-hour Emergency Preparedness Response Plan design sessions between Paris-based facilitators and approximately thirty humanitarian aid workers in Yemen, gaining invaluable insights into actual training delivery dynamics, facilitation techniques, participant engagement patterns, and the challenges of supporting distributed teams across unstable communication infrastructure. This immersive approach to requirements discovery ensures that technical development serves genuine operational needs rather than implementing capabilities that seem promising in abstract discussions but deliver limited practical value in deployment contexts.

Immersive Prototyping Sessions

The requirements process includes collaborative sessions in Mozilla Hubs virtual environments where Nuwa and Action Contre la Faim teams meet in rudimentary emergency coordination spaces experiencing spatial collaboration firsthand. These immersive sessions enable Action Contre la Faim personnel (many experiencing virtual reality for the first time) to viscerally comprehend what immersive training might offer their organisation rather than relying on abstract descriptions that prove difficult to evaluate without experiential reference points. The design team demonstrates wireframed mock-ups illustrating potential virtual environments (coordination offices, field locations, community spaces), showcases early proof-of-concept features including presentation systems, 3D object interaction, and six-degrees-of-freedom navigation, and engages two-way dialogue about how these capabilities could support specific learning objectives and operational requirements. This hands-on approach proves invaluable for moving conversations from speculative "could this work?" discussions to concrete "here's specifically what we need" design sessions grounded in tangible experience rather than imagined possibilities.

Early AI Avatar Demonstration

The team has developed early demonstration of "Mentor Maud", an AI-powered avatar concept showcasing how large language model technology grounded in organisational documentation could enable interactive knowledge transfer and question-answering without requiring live facilitator availability for basic conceptual briefings. This tangible prototype generated substantial enthusiasm from Action Contre la Faim stakeholders, validating the value of conversational AI capabilities and illustrating how abstract technical descriptions ("we could integrate LLM dialogue") transform into concrete experiential understanding ("I just had a realistic conversation with a virtual emergency coordinator") that enables informed decision-making about capability priorities. The demonstration approach (rapid prototyping of key concepts enabling stakeholder interaction with working implementations) accelerates requirements validation compared to specification documents that often generate only superficial approval that evaporates upon encountering actual functionality.

Requirements Outcomes and Next Steps

The collaborative requirements gathering process is producing detailed pilot specifications including learning matrices defining objectives, tasks, required environments, communication needs, information requirements, user interface specifications, and evaluation criteria for each emergency management cycle phase. Weekly requirements sessions alternate between conventional video conferencing and immersive virtual world meetings, with both modalities contributing distinct value: videoconferencing enables efficient information exchange and document review whilst immersive sessions enable experiential understanding and spatial collaboration exploration. The requirements definition will inform Phase 2 implementation priorities, ensuring development focuses on capabilities with validated operational value rather than pursuing technically impressive features without clear training outcome contributions.

Organisational Learning

The process demonstrates that effective requirements gathering for specialised domain applications requires substantial investment in understanding operational contexts, user workflows, and institutional constraints before proposing technical solutions. The methodology of observing actual training delivery, participating in existing learning modules, and engaging stakeholders in immersive prototyping proves far more effective than conventional requirements interviews or specification document exchanges that often miss implicit needs and unstated assumptions. This user-centric approach, whilst time-intensive compared to rapid specification processes, generates alignment between development priorities and genuine user needs that prevents wasted effort on capabilities delivering minimal operational benefits despite technical sophistication.