Mira Tarot

A question today. A space to return to tomorrow.

I've been reading tarot for years, for myself and for friends. Mira began with a simple question: how could a reading feel personal when a human reader isn't available? It has grown into an iOS app combining tarot, AI interpretations, personal wishes, and a lightweight birth chart.

I lead the product definition, interaction and visual design, and AI-assisted development. This case study captures a working version: the core flows are implemented, while testing, UI refinement, and information hierarchy remain ongoing.

Why a dedicated app, when you can ask AI?

My focus shifted from generating an answer to creating continuity around a question. Drawing cards offers a deliberate starting ritual. Saved readings preserve context. Wishes give longer-term interests a place to live beyond a single session.

The hypothesis is that these connections can make reflection easier to return to. That is a design intention, not a proven retention result; broader user validation is still ahead.

01 / A small daily ritual

Home brings together a daily card, a short reflection, and a small action. The card stays consistent for the day by default, with a quieter redraw option for someone who wants another perspective. Signed-in users can also see their daily free tarot AI allowance and remaining AI reading credits.

02 / A reading organized around the question

Users ask a question, draw cards, and choose whether to request an AI interpretation. The default three-card reading is a freeform synthesis: it does not automatically treat the middle card or a Major Arcana card as the main card. Guided spreads retain their explicitly defined positions.

I am refining the AI instructions to answer the actual question, connect the cards, and offer a concrete observation or action. Warmth should come from relevance, not generic reassurance or invented personal details. Response consistency is still being tested.

03 / Wishes that connect individual readings

A wish becomes a star. Users choose its color and emoji, arrange it by dragging, and tap to revisit its details. Each wish offers three suggested tarot questions, creating a starting point grounded in something the user already cares about.

An optional birth chart adds Sun, Moon, and rising signs. These can inform wish-specific guidance, while the raw birth details remain on the device. This is a lightweight reflection feature, not a professional astrology suite.

A lighter visual direction — still finding its hierarchy

The interface has evolved from deep purple surfaces toward pale lavender, deeper purple text, and selective champagne-gold accents. The wish space keeps a night-sky atmosphere, while birth-chart details and reading packs use a slightly deeper lavender.

The palette now feels more consistent, but the screens can still feel too flat. Cards, supporting information, and actions sometimes carry similar visual weight. My next priority is a clearer reading order, not simply adding more decoration.

Built, tested, and still evolving

I designed and built Mira in SwiftUI, using AI-assisted development to iterate on working flows and test them in the simulator and on a personal iPhone. The current build includes daily cards, tarot AI interpretations, saved history, follow-up questions, wish stars, basic birth-chart calculations, and profile customization.

Email and Google sign-in connect to the existing account service. The purchase interface is present, but payment processing and complete transaction management remain unfinished. Readings, wishes, and birth details currently rely primarily on local storage; cross-device synchronization is still ahead.

The app has not formally launched. Current evidence comes from local checks and hands-on testing, with no growth, conversion, or retention results to report. The next phase combines broader validation with improvements to clarity and reliability.

Also built: Mira Tarot Web

Alongside the iOS app, I built a gesture-controlled tarot website where an open-palm gesture draws cards through the device camera. It explores a more physical way to begin the same ritual.

Try the gesture demo →

Reflection

Building more features did not automatically create a more coherent experience. This iteration brought the focus back to the user’s question: how they begin, what makes an answer feel attentive, and what is worth returning to. Mira is now moving from functional flows toward a clearer, more dependable experience.