24/7 Mental HealthSupport

Rafi provides immediate emotional support, tracks wellbeing trends, identifies students at risk, and connects them with human counselors — exactly when it matters.

24/7Always available
English & SwahiliBilingual support
100%Crisis escalation rate
Rafi
RafiWellbeing Companion
Habari! I'm here whenever you need someone to talk to. How are you feeling today?
Neutral
Exams are in two weeks and I haven't been sleeping. I feel so overwhelmed.
That sounds really exhausting — carrying exam pressure while running on little sleep. You're not alone. What's weighing on you most right now?
Stress detectedSleep concern
The problem

People are struggling in silence

Millions of People live with anxiety, depression, burnout, loneliness, and financial stress every day. Yet most never seek help — not because they don't need it, but because the barriers are too high.

Mental health services are overwhelmed and understaffed. People wait weeks for appointments while their struggles compound. Early intervention is nearly impossible at scale.

Rafi changes that — by being available the moment a person needs support, not a fortnight later.

1 in 4

Kenyans will experience a mental illness at some point in their lives

75%

of Kenyans have no access to mental health services

3–4 wks

average wait time for a mental health appointment in Kenya

1 : 500,000

typical psychiatrist-to-population ratio in Kenya

What is Rafi?

Proactive wellbeing, not reactive care

Rafi is an AI-powered mental wellbeing platform designed specifically for universities. People receive immediate emotional support any time of day, while mental health professionals gain the visibility they need to intervene before a crisis escalates.

It's not a chatbot. It's a clinically-informed support system that understands context, detects risk, and knows when to bring in a human.

Emotionally intelligentCulturally awareEvidence-basedUniversity-integratedPrivacy-firstBilingual (EN / SW)

Person wellbeing snapshot

Mood improving+18% this week
Sessions this week3 sessions
Risk levelLow
Top stressorAcademic exams
Core features

Everything a person needs. Everything a mental health professional wants.

AI counseling companion

People access supportive, judgment-free conversations anytime. Responds with empathy grounded in evidence-based counseling frameworks.

  • Available 24/7 — no appointments needed
  • English and Swahili language support
  • Context-aware, personalized responses
  • CBT and mindfulness-based techniques

Mood tracking & wellbeing monitoring

People log moods and visualize emotional trends over time — building self-awareness and spotting patterns before they become problems.

Calm
72%
Stressed
44%
Anxious
21%

Early risk detection

Rafi continuously analyzes conversations for signs of distress — catching struggling people early, before a crisis develops.

  • Multi-signal risk classification
  • Detects subtle distress indicators
  • Reduces time to intervention significantly
  • Prioritizes urgent cases automatically

Human mental health professional escalation

AI support never replaces professional care. When risk is detected, the transition to a human mental health professional is smooth, warm, and immediate.

  • Three-tier escalation system
  • Mental health professional receives full session context
  • Person-consented handoff process
  • Crisis resources surfaced instantly

Person reflection & progress

People can review past sessions, track their emotional growth over time, and recognize recurring patterns — turning every conversation into a step toward long-term wellbeing.

  • Session history and summaries
  • Emotional trend visualization
  • Personalized coping strategy library

Your progress this month

Wk 1Wk 2Wk 3Wk 4
↑ Wellbeing score improving — keep it up
Built for everyone on campus

Tools for people. Visibility for professionals. Insights for institutions.

For professionals

Mental Health Professional Dashboard

Give mental health professionals tools, not extra work. The dashboard surfaces the people who need them most, with full context — so every session starts informed.

Flagged person monitoring

See at-risk people ranked by urgency before they reach out

Conversation review

Access session transcripts (with person consent) to prepare for handoffs

Intervention scheduling

Book follow-ups directly from the dashboard without leaving the platform

For universities

Institutional wellbeing insights

Administrators gain population-level visibility into mental health trends — enabling smarter resource allocation and proactive policy-making.

Campus trend analytics

Anonymised wellbeing data across faculties, year groups, and campuses

Risk trend tracking

Identify high-stress periods like exams before they peak

Resource planning support

Data-driven recommendations for counseling staffing needs

Why Rafi is different

Multi-layer mental health intelligence

Instead of simple chatbot responses, Rafi understands what's underneath — the emotion, the cause, the right approach — and acts on it.

Understands emotions

Emotion classification reads beyond words — tone, phrasing, and emotional context are all considered before a response is generated.

Identifies root causes

A causal reasoning layer identifies underlying stressors — academic, financial, social — so responses address the actual problem, not just the surface symptom.

Selects the right strategy

A planning module chooses from evidence-based counseling strategies — CBT, motivational interviewing, psychoeducation — matched to each person's situation.

Responds with empathy

Responses are warm, human, and culturally appropriate — not clinical or robotic. Bilingual in English and Swahili.

Detects crisis signals

A dedicated safety supervision layer watches every conversation for crisis indicators and escalates immediately — no delay, no missed signals.

Built for Kenya

Trained on the Kenyan university experience — local academic calendar, cultural context, and specific stressors that generic global products miss entirely.

94%Sessions resolved by AI without escalation
< 4 minAverage mental health professional response on escalations
12+Partner universities across Kenya
100%Crisis cases escalated to human mental health professionals
Safety first

Escalation when it matters most

Rafi knows its limits. A smart triage system ensures no person in crisis is left without human support.

Level 1 — Routine

AI handles independently

Everyday stress, academic pressure, mild anxiety. The AI provides coping strategies, guided exercises, and resource links. Session notes logged for trend monitoring.

Level 2 — Moderate risk

Mental health professional alerted

Persistent distress, self-worth concerns, social isolation. AI flags the session and notifies the mental health professional dashboard. Warm handoff offered to the person.

Level 3 — Crisis

Immediate professional connect

Suicidal ideation or self-harm signals detected. Mental health professional is immediately paged. Emergency contacts and Befrienders Kenya hotline presented. Session fully transferred.

Mental Health Professional Dashboard

Mental health professionals stay in control

A real-time dashboard shows session risk levels across the body, flagged cases, and complete conversation context — so mental health professionals can act fast with full information.

94%Resolved by AI
4 minAvg mental health professional response
100%Crisis cases escalated
Privacy & safety

Built for trust from the ground up

Mental health data is deeply personal. Rafi treats it that way — with end-to-end encryption, strict access controls, and full compliance with Kenyan law.

End-to-end encryption

Every conversation is encrypted in transit and at rest. No third party can access student session data.

Consent-driven access

Counselors can only view session content with explicit student consent. Role-based access enforced at every level.

Kenya DPA 2019 compliant

Fully compliant with Kenya's Data Protection Act 2019. Institutional data processed and stored locally where required.

Secure cloud infrastructure

Built on enterprise-grade infrastructure with redundancy, monitoring, and penetration-tested security protocols.

Anonymous option

Students can use Rafi anonymously. Zero stigma — no institutional login required to start a conversation.

Privacy-first design

Data minimization by default. Only what's needed for care is stored. Students can delete their data at any time.

Technical foundation

Powered by advanced AI

A multi-layer architecture designed specifically for the nuance and sensitivity required in mental health support.

Core AI pipeline

01Emotion Classifier
→
02Causal Analysis
→
03Strategy Planner
→
04Response Generator
Safety supervision layer (always active)
Emotion classification engine
Risk assessment models
Facial emotion recognition (FER)
Speech emotion recognition (SER)
Multi-agent reasoning architecture
Safety supervision layer

Trusted by Kenyan universities

JKUAT
University of Nairobi
Kenyatta University
Strathmore University
Moi University
Multimedia University
Get Started

Transform wellbeing across your community

Rafi helps universities move from reactive counseling to proactive, always-on people support — starting with your next semester.

Free pilot for partner universities
User data stays on-premise
Live in under 2 weeks
Dedicated onboarding support

Get in touch