Sex Therapy Pro

The Problem

Sexual wellness apps often feel clinical or gimmicky. People need a private space that can listen without judgment, turn goals into a day-to-day plan, and offer exercises that feel human — not a generic chatbot with a lavender theme.

What I Built

I owned the Expo / React Native frontend for Sex Therapy Pro: an AI sexual-wellness companion with a personalized daily plan, chat with Eros (AI therapist powered by AWS Bedrock), interactive desire/body mapping, and couple-oriented exercises such as The 3 Minute Game.

Major Features

• Personalized plan — Daily task progress (e.g. sensate touch, desire mapping, homework) with a calm home dashboard and bottom nav.
• Chat with Eros — Conversational intake (“Improve desire/arousal”), follow-ups, free-form “Ask me anything,” and an Analyze action to turn chat into guidance via Bedrock.
• Topic & session hub — Choose therapy themes (attachment, boundaries, emotional expression) and revisit recent AI sessions.
• Onboarding goals — Bubble selection for how you want to feel after using the app (connected, knowledgeable, less ashamed, etc.).
• Guided exercises — Sensory foreplay body map, desire mapping, and timed couple games introduced with clear start screens.

Key Challenges

Sensitive UX without feeling medical or casual. Soft gradients, illustration, and copy had to invite trust while keeping hierarchy clear for plan → chat → exercise.

AI chat as product, not a text box. Intake, Analyze, and recent sessions needed to feel like a therapist workflow, with Bedrock responses wired into a calm mobile conversation UI.

Interactive tools on mobile. Body-map selection and multi-select goal bubbles had to stay readable and tappable without cluttering the pastel layout.

Results

Sex Therapy Pro shipped as a mobile AI wellness frontend: personalized plans, Eros chat, and guided intimacy exercises in one Expo client.

Stack Decisions

Expo / React Native for iOS and Android from one codebase. AWS Bedrock powers Eros and Analyze so conversational guidance stays on a managed model path; the plan and exercise screens turn that output into concrete daily actions instead of leaving users in an endless chat thread.