Gnothiai

Implements a multi-turn dialogue system where an LLM chatbot analyzes user journal entries in real-time and generates contextually-aware follow-up questions designed to deepen reflection. The system maintains conversation state across sessions, allowing the bot to reference previous entries and build on prior insights. Uses prompt engineering to guide users toward deeper self-discovery rather than surface-level responses, with the chatbot acting as a Socratic coach that asks clarifying questions based on detected emotional themes or unresolved tensions in the user's writing.

Provides a library of guided meditation sessions organized by duration, theme (stress relief, sleep, focus), and difficulty level. Sessions are delivered as pre-recorded audio with optional visual progress indicators and session metadata (duration, instructor, technique type). The system likely uses a content management backend to catalog sessions and a streaming audio player to deliver content with offline caching support. Sessions may include biometric integration hooks (e.g., heart rate monitoring) but core functionality is audio playback with minimal interactive elements.

Captures user mood states, physical wellness metrics (sleep, exercise, nutrition), and emotional patterns through structured input (mood tags, rating scales) and correlates them with journal entries and meditation sessions over time. The system stores time-series data and generates trend visualizations (mood over weeks/months, correlation between meditation frequency and reported stress levels). Uses simple statistical aggregation to identify patterns (e.g., 'you report better sleep on days you meditate') without requiring complex ML—primarily a data collection and visualization layer.

Uses NLP or LLM-based analysis to parse journal entries and automatically generate tailored reflection prompts that target unresolved themes, emotional gaps, or areas of potential growth. The system identifies key topics (relationships, work stress, health concerns) and generates follow-up prompts designed to deepen exploration of those specific areas. Prompts are delivered either immediately after entry submission or as part of a daily/weekly reflection digest, with the option for users to accept or dismiss suggestions.

Maintains a persistent store of user journal entries, meditation sessions, mood logs, and chatbot conversations, allowing the AI to reference past interactions and build a coherent narrative of the user's growth journey. The system implements a retrieval mechanism (likely vector embeddings or keyword search) to surface relevant past entries when the user starts a new conversation, enabling the chatbot to say things like 'Last month you mentioned struggling with X—how is that going now?' This requires a database schema that links entries, conversations, and metadata, plus a retrieval pipeline that identifies contextually relevant history.

Implements a freemium pricing model where core journaling and meditation features are available without payment, while premium tiers unlock advanced features (likely: unlimited AI conversations, advanced analytics, premium meditation content, offline access). The system uses account-level feature flags or subscription status checks to gate functionality at runtime, allowing free users to experience the product's core value proposition before deciding to upgrade. Monetization likely relies on conversion of engaged free users to paid tiers rather than aggressive paywalls.

Presents a consolidated view of the user's wellness activities across journaling, meditation, and mood tracking in a single dashboard interface. The dashboard likely displays widgets showing recent journal entries, upcoming meditation sessions, mood trends, and AI coaching insights, with the ability to drill down into each section. This requires a data aggregation layer that pulls from multiple subsystems (journal database, meditation library, mood tracker, chatbot logs) and presents them in a unified UX without requiring the user to navigate between separate screens.

Extends the chatbot beyond simple Q&A to provide ongoing coaching through multi-turn conversations where the AI offers guidance, accountability, and encouragement based on the user's journal entries and wellness goals. The coaching system uses conversational patterns (motivational interviewing, Socratic questioning, validation) to help users identify barriers to change and develop action plans. The AI maintains a coaching context across sessions, remembering previous goals and progress, and can proactively check in on commitments the user made in prior conversations.