Building systems that scale and improve care
The future of mental healthcare depends on systems that can both deliver and improve care at scale. We are building that system—combining infrastructure for delivery with intelligence that makes care better over time.
Technology that makes scale possible
Scaling mental healthcare requires more than delivering care—it requires systems that can maintain quality and improve over time. We are building that infrastructure through two connected systems: one that enables delivery, and one that strengthens it through continuous learning.
shamiriOS: The system that powers care delivery
shamiriOS is the operating system for care delivery. It coordinates providers, clients, and sessions—making it possible to deliver care consistently across sites, organisations, and contexts. It is designed for real-world conditions, including distributed teams and low-resource environments.
Learn more about shamiriOS ↗shamiriAI: The system that powers quality and learning
shamiriAI is the intelligence layer. It analyzes care delivery to support supervision, measure fidelity, and generate the feedback loops needed to improve quality. As the system grows, it helps ensure that care does not degrade—but improves.
Learn more about shamiriAI ↗Featured research
shamiriAI for Supervision Pilot Validation Study
We developed and pilot-validated shamiriAI, an automated fidelity monitoring tool for lay-delivered mental health interventions, embedded within the Shamiri school-based mental health program in Kenya. shamiriAI demonstrates technically feasible multilingual ASR and a coherent, interpretable reliability profile. Underperformance was concentrated on holistic inferential dimensions — particularly Required Contents and Clarity — while structured detection tasks already approach operational utility. The underperformance pattern reflects diagnosable misalignments in rubric interpretation and prompt design, with clear engineering solutions identified. These findings provide the foundational validation evidence and dimension-specific diagnostics needed to guide the development of AI-augmented supervision for lay-delivered adolescent mental health programs in sub-Saharan Africa and other multilingual settings.
Randomized controlled trial of the Shamiri intervention
A randomized trial of 413 Kenyan adolescents found that Shamiri, a 4-week lay-provider-delivered intervention, produced greater reductions in depression and anxiety symptoms at posttreatment (d = 0.35 for depression, d = 0.37 for anxiety) and sustained effects through 7-month follow-up.
Pilot randomized trial of Shamiri in Kibera
A pilot RCT of 51 Kenyan adolescents in an urban slum found that the Shamiri intervention produced greater reductions in depression (d = .32) and anxiety (d = .54) compared to study skills control, and improved academic performance.