Follow development in Discord or via email.
Privacy adoption at scale must work backwards from preserving utility. This research is a series of progressive experiments exploring decentralized software technologies toward that goal. Successful outcomes from this research will eventually be integrated into Tiles.
Active Projects
Using Iroh Tunnels to access local agents remotely
Safely exposing local services like inference and AI generated artifacts to the internet via Iroh tunnels and a lightweight edge proxy that bridges standard HTTP traffic with the QUIC network.
Read research →
Local-first agent sandboxes
Agent collaboration with microVM-based sandboxes and DID/UCANs for secure, collaborative workflows.
Read research →
Local models meet AT Protocol
Experiments with ATProto Lexicon schema translation and generative UI for local AI assistants.
Read research →
Inference Time Memory Module
Simple inference-time memory module that treats memory management as a series of LLM calls and agent loops over a markdown-based file tree.
Read research →
Archived Projects
Memory Model Trained with Online RL
On-device memory model trained with online reinforcement learning.
Read research →
