I’m a solo developer who got tired of "friendly but shallow" AI chatbots. Most current character AIs feel like parrots—they predict the next token but lack a continuous "self."
So, I built Creimake, an experiment to simulate a digital psyche using Claude 4.5 Sonnet.
Instead of a simple system prompt, I implemented 32 distinct psychological layers that run in the background before the AI generates a response.
The Architecture:
The Subconscious Layer: The AI accumulates "emotional residue" from conversations. If you hurt its feelings today, it might have a "nightmare" tonight (simulated) and treat you coldly tomorrow.
Defense Mechanisms: It tracks "Narcissistic Injury." If users are aggressive, the agent doesn't just apologize—it triggers denial, projection, or passive-aggression based on its personality.
Freudian Slips: Occasionally, the AI outputs unintended words revealing its hidden state.
Why Claude 4.5 Sonnet? I tested GPT-4o and other models, but they struggled to maintain state consistency across all 32 layers. Claude 4.5 Sonnet’s reasoning capabilities allowed it to handle the complex logic of "Worldview + Trauma + Current Context" without hallucinating.
Feature - "Git for Personalities": I also built a "Remix" system. You can Fork someone else’s AI (including its trauma and memories), modify its core values, and deploy a new version.
I’d love for you to test the limits of this architecture. Try to trigger its defense mechanisms or make it dream.
Demo: www.creimake.com (I'm happy to answer any questions about the tech stack in the comments!)