Plasticity-AI is a one-person research studio exploring how AI systems can be made more responsive, more reliable, and more respectful of the human experience.
Our work is centered around real-time, local-first tools that prioritize integrity over scale — systems that reset cleanly and adapt under pressure.
We’re not trying to build artificial general intelligence in the traditional sense. We’re focused on something more niche:
making machine behavior feel grounded, explainable, and alive.
At the core of our work is Evelyn, a voice-native assistant designed around clean loops, memory anchoring, and forensic fallback handling. But Evelyn is also a proving ground — a way to test ideas about how cognition, context, and trust might function in machine systems.
She was built with particular foundations:
Evelyn runs entirely on local hardware — not in the cloud — using open-source LLMs, Whisper transcription, custom memory architecture, and an orchestration layer that treats latency, fallbacks, and reset integrity as sacred.
We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.