Kyndras

A virtual advisor site that measures AI usage and converts tokens to credits. Highly trained virtual advisors across widely accepted wellness modalities.

The problem

Access to quality wellness guidance is expensive and inconsistent. Therapy sessions cost $150-300/hr. Wellness coaches vary wildly in quality. Most people who need support don't seek it because of cost, scheduling, or stigma.

What we built

AI-powered virtual advisors trained in specific wellness modalities: cognitive behavioral techniques, mindfulness, journaling guidance, and goal setting. Each advisor has a distinct personality and approach. Users purchase credits that convert from AI token usage, making pricing transparent and predictable.

Built in 3 days with a novel token-to-credit metering system that lets users understand exactly what they're paying for.

Screenshots

Features

What it does

Virtual wellness advisors

AI personas trained in CBT, mindfulness, journaling, and goal setting.

Token-to-credit metering

Transparent usage tracking that converts AI tokens to user-facing credits.

Session history

Full conversation history with mood tracking and progress notes.

Advisor personalities

Each advisor has a distinct name, specialty, and communication style.

Safety boundaries

System knows when to recommend professional help and never oversteps.

Process

How we built it

Researched wellness modalities

Studied CBT, mindfulness, motivational interviewing, and journaling frameworks to build accurate system prompts.

Designed the advisor personalities

Each advisor has a name, specialty, communication style, and therapeutic approach. Not generic chatbots.

Built the token metering system

Tracks AI token usage per conversation and converts to a credit system users can understand and budget.

Created the conversation interface

Chat-style interface with session history, mood tracking, and progress notes.

Tested with wellness professionals

Had licensed therapists review conversations for accuracy, safety, and appropriate boundaries.

Reflections

What we took away from this project.

What went wrong

Early versions were too clinical. Users wanted warmth and personality, not a textbook. We had to significantly adjust the advisor tones.

What went right

The credit system built immediate trust. Users could see exactly how their credits were being used, which reduced churn significantly.

What we learned

AI wellness tools need clear boundaries. The system must know when to recommend professional help and never pretend to be a replacement for it.

Want something like this for your business?

Tell us what you're working on. We'll tell you how fast we can build it.