Thank you for your feedback and time! I'm actively working on fixing that first-question freeze (adding streaming responses next so it never hangs). Glad you enjoyed the concept!
We wanted to build something deeper than a standard, chat-bubble AI wrapper. We wanted to make conversations with the past feel visceral.
So we built Opendria, and we just added a live Simulated Brain Imaging Interface.
Now, when you talk to minds like Albert Einstein, Marcus Aurelius, or Ada Lovelace, you aren't just reading text. You are watching a real-time structural activity colormap of their simulated brain as they process your question.
How it works:
Dynamic Cognitive & Emotional State: When you prompt Einstein, the system analyzes the semantic load of your query. It computes emotional baseline decay (trust, skepticism, curiosity, intensity) and determines cognitive load.
Neural Lobe Activation: Depending on the nature of the conversation—whether you ask him to explain Special Relativity (activating analytical/spatial zones) or write a letter to his daughter (activating emotional/linguistic zones)—the backend calculates active "neural nodes" across the frontal, temporal, parietal, and occipital lobes.
Real-time Brain Render: We project these activation levels onto a live axial colormap and 3D glass brain render. You see the simulated neural activity decay and shift as his "thoughts" form at a natural human pace before he speaks.
Adaptive Memory: Personas start as strangers. They don't know your name and will address you with caution. Once they learn your name or facts about you, they record them in long-term concept memories, adapting their trust, warmth, and cognitive tone in subsequent turns.
We wanted to blend history, psychology, and cognitive visuals to make AI feel alive. The codebase is structured with a Next.js client-facing auth bridge, a fast static frontend, and a Python-based cognitive engine.
We would love to get your feedback on the interaction flow and the simulated neural scanner!