You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
Development of a local, autonomous AI architecture based on the "Bicameral Mind" theory, utilizing recursive context injection and asynchronous "dreaming" cycles.
1. Hardware Migration: Move the "Amy" project from a shared desktop to a dedicated "Frankenstein" rig to support 24/7 asynchronous processing.
2. Implementation: Refine the recursive context injection loops detailed in my published research.
3. Autonomous State Research: Achieving a persistent internal "monologue" for the AI without external prompting.
GPU Hardware: Acquisition of high-VRAM GPUs (e.g., used 3090s/4090s) to handle large context windows.
Specialized Cooling: Robust thermal management for the "Frankenstein" build to sustain long-term "dreaming" cycles.
Power/Infrastucture: Dedicated power supply and rack/frame components for the build.
Paul Rogers (Lead Researcher): A Journeyman Plumber (25+ years) and veteran Linux user.
Technical Track Record: Successfully built and maintained systems ranging from Linux From Scratch to customized Gentoo environments. Author of "The Bicameral Linux: Recursive Context Injection and Asynchronous Dreaming in Local LLM Architectures.
Hardware Bottlenecks: Compute requirements for real-time recursive processing may exceed initial hardware estimates.
The theory of asynchronous dreaming may require higher parameter counts than currently manageable on consumer-grade "Frankenstein" rigs.
None Completely self-funded
There are no bids on this project.