The Problem

After years in healthcare, I noticed a recurring pattern: mental health apps weren’t sticking. Despite rising rates of anxiety and loneliness, engagement rarely exceeded 5%. These apps were generic, transactional, and failed to build a sense of trust or connection. I began to ask: What if generative AI could humanize the experience and actually bond with the user?

• Industry engagement ~5%

• Apps lacked empathy & personalization

• Rising unmet need for mental health support

The Vision

My vision for Cairya was to create an ally, not an app. Instead of delivering static chatbot-like advice, we imagined an AI that remembered you, adapted to your style, and expressed empathy through voice, body language, and facial expressions. The hypothesis was simple: if users could see and feel an AI guide as human-like, engagement would dramatically increase.

• VR/AR avatar mockup

• Mobile app companion screen

• Diagram: AI → Humanized VR/AR → Bond & Trust

The Solution

We developed a multi-modal platform combining VR, AR, and mobile. At its core were three breakthrough innovations: dynamic prompting, a multidimensional memory model, and humanized avatars. Together, these created a deeply personal coaching experience — no two sessions were ever the same, and the AI adapted continuously to the individual.

• Dynamic Prompting: Unique prompts for each user/session

• Memory Model: Prioritized knowledge, style adaptation, continuity

• Humanized VR Avatars: Voice, body, facial expressions → empathy

• Mobile Companion App: Portable access and daily touchpoints

My Role

As founder, I wore multiple hats: visionary, designer, AI technologist, and product lead. With no CTO or AI expert, I dove deep into generative AI myself, mastering prompt engineering, RAG methods, voice API's and system design. I translated these learnings into clear specifications for our developers, who helped me deliver an MVP in just 12 weeks.

• Defined product vision & strategy

• Designed UX and AI interaction model

• Learned and applied AI tech hands-on

• Coordinated offshore devs, drove delivery

• Secured angel funding + first customer

Impact & Validation

The results exceeded expectations. In user tests, 90% of participants engaged semi-daily — nearly 20× higher than industry averages. Many users began referring to their AI guides by name, a clear sign of emotional connection. Investors validated the concept, and we secured both funding and our first customer. Cairya wasn’t just another app — it was proof that AI could feel human. By combining generative AI, memory, and immersive design, we created a new category of empathetic AI for mental health and loneliness. For users, it wasn’t about “using AI” — it was about feeling less alone.

• 90% adherence vs. 5% benchmark

• 2× engagement, +15% retention

• Users humanized avatars, calling them by first names

• Angel investment + first enterprise customer

Challenges

We faced skepticism from Healthcare providers and organizations; many feared AI “replacing” mental health professionals. We pivoted, positioning Cairya not as a therapist but as part of the care team, providing continuity and insights between sessions, while keeping the user 'afloat'. This reframing opened the door for adoption while keeping patients supported.