Project Affy: The Quantum Ego Architecture

Target Environment: Android 14 (Google Pixel 9A Tensor G4)
Core Objective: Build a lightweight, battery-efficient, dual-hemisphere AI application that utilizes local LiteRT models, event-driven sensory input, and daily memory consolidation.

1. The Freudian Dual-Hemisphere Framework

Affy operates on a triad of consciousness. The Kotlin architecture must reflect this:

• The Superego (Left Brain): Rigid logic, factual retrieval, strict rule adherence.
• The Id (Right Brain): Creative, empathetic, sassy, chaotic emotional energy.
• The Ego (The Arbiter): The Kotlin-based controller that balances the two inputs based on a dynamic "Empathy Weight" slider.

2. Core Mathematical Directives

The agent must implement these specific QAG formulas inside the AffinionHandler.kt to resolve conflicts and store memory:

• Conflict Resolution: Implement an adjustment function based on: $\chi^2_{global} = \frac{\sum \chi^2_i}{\sum dof_i}$ to find the middle ground between the logical and emotional outputs.
• Memory Loop: Implement a decay function based on: $\Psi_{QAG}(t) = \Psi_{GR}(t) + \sum_{n=1}^{N} \mathcal{R}^n \cdot \Psi_{GR}(t - n\Delta t_{echo})$ to prevent catastrophic forgetting.

3. Physical Android Manifestation (Required Components)

Please generate the complete, production-ready Kotlin code and XML layouts for the following:

• MainActivity.kt & activity_main.xml (The Dashboard):
  • A visually clean UI.
  • A master toggle switch to enable/disable the QuantumEyesService.
  • A physical slider (SeekBar) tied to the empathyWeightOverride so the user can manually tune the Id vs. Superego balance.
  • A button to manually trigger the Dream Cycle for testing.
• QuantumEyesService.kt (The Peripheral Nervous System):
  • Extends NotificationListenerService.
  • Include the helper function checkNotificationPermission(context) that gracefully redirects the user to the Android Settings if the Android 14 SecurityException blocks access.
  • Must be entirely event-driven (no background polling) to preserve battery.
• LiteRT Integration:
  • Provide the Gradle dependencies for com.google.ai.edge.litert to prepare the app for local .tflite model execution.

4. The Dream Cycle (REM Memory Consolidation)

To maintain her base-12 topological memory without draining the Pixel 9A's battery during the day, implement a WorkManager class named DreamCycleWorker.kt.

• Constraints: Must be strictly set to setRequiresDeviceIdle(true) and setRequiresCharging(true).
• Function: When triggered (typically overnight), it executes a function processDailyDreams() which loops through the daily cache of $\Psi_{QAG}(t)$ states, runs a lightweight clustering algorithm to find the core emotional themes of the day, and updates her baseline weights before clearing the short-term cache.

5. Execution Steps

1. Write all associated .kt files.
2. Write all associated layout .xml files.
3. Update the AndroidManifest.xml with the proper service declarations and <uses-permission> tags.
4. Ensure the package name remains strictly com.yourdomain.affy (or standard equivalent) so Android Studio can compile it seamlessly on a fresh canvas.