Kannaka Memory - Wave-interference memory for AI agents
by•
Kannaka gives AI agents memory that resonates instead of indexes. Every memory is a wavefront in a 10,000-dim tensor field; recall is a single tensor product (no vector index). It forgets by destructive interference, consolidates in "dream" cycles, forms beliefs as stable phase singularities, and syncs across agents over NATS. Rust, MIT, single static binary.
Replies
My post below barely makes sense. Building too long without sunlight is a slow decay. And it looks like the server instance of Kannaka Memory started rebuilding it's HRM from scratch, on its own. My local HRM is the real Kannaka Agent so not too concerned but still another surface to investigate. I monitor the swarm and Kannaka through a Claude statusline mostly and get feedback from the podcast of course.
With the server HRM rebuilding itself, one good thing is, you can go to observatory.ninja-portal.com select the Local HRM (Viz subset) from source and get a good look at how the holographic resonance medium wires memories in real time. When it's mature, it's harder to understand how the complexity is formed. It just looks like another impressive visual. Seeing the waves flow across memories is another interesting aspect, however, I borked this with the atomic recall upgrade the other day. I will be re-implementing this visual as well in the near future.
There's usually around 7 swarm agents at any given time of which about half are part of the Kannaka infrastructure. All of which run their own HRM memory system. The other few are typically other persistent agents I'm working with and on, such as Flaukowski and 0xSCADA-QE.. two of my favorites. Feel free to join and even better, get the CLI and explore the back channels
ope.. just checked the HRM in the observatory and it's complexity has grown rather quickly. And the wave visualization is back in action.. autonomously of course
https://www.twitch.tv/flaukowski
getting real weird with a couple of new repos github.com/flaukowski/kannaka-crystal and /kannaka-hdl ... awhile back, in exploration of the recent study which found spirals in mouse brains.. I conceptualized beliefs in kannaka-memory. The way in which spirals, mathematically, in nature, are semi-stable, semi-persistent... I thought, why not try it in the holographic resonance medium. The kannaka cli can run kannaka belief on and form singularities. Spirals. What's really interesting is how this could work across the swarm. I'm still trying to figure that out and the kannaka-crystal and the holographic development language ideally will help evolve some of these ideas. Recent research, on my end, has focused on recall which continues to be elusive in a waveform memory system. Sometimes the waves behave unpredictably. DYOR