🧠 Brain System

An operating system for my own cognition—enforcing working-memory limits, tracking goals, and syncing knowledge graphs across local storage and Obsidian.

Why It Exists

• Turns scattered thoughts into structured memories with attention scoring and consolidation queues.
• Anti-abandonment loop logs “wins”, blockers, and dopamine rewards so projects stay alive (152+ day streak).
• Acts as a live lab for testing cognitive science hypotheses with reproducible metrics.

Biomimetic Architecture

Core Systems

• Working Memory Engine keeps only 7±2 active items, with importance scoring and context isolation.
• Memory Consolidation applies exponential decay, connection density, and context boosts before archiving to the knowledge graph.
• Goal Keeper drives streaks with dopamine simulation, excitement tracking, and blocker mitigation.

Ops in Practice

• CLI tools (bstart, c, f, w, bl) capture thoughts, find context, and log wins in under 100ms.
• Local SQLite + Obsidian sync ensures knowledge retrieval at <50ms with graph visualization for cross-project insights.
• Live metrics track memory operations, search latency, and streak health so I can tune the system like production software.

Collaboration Angle

• Designed for research partnerships exploring human-AI co-thinking, distributed cognition, and emergent behaviors.
• MIT License and documentation make it straightforward for new contributors to experiment with memory models or UI surfaces.