The first signalsfrom a real field.
One published demonstration from a real Rhizome run, followed by two clearly labeled reports in preparation. The published signal ships with source, logs, and honest notes on what works and what doesn't.
Autonomous build
A working program, built end to end.
A three-agent run took a single specification and produced rq — a working JSON expression language with a REPL. Product code authored by agents, with the full commit history, evidence bundle, and an explicit claim boundary published.
Autonomous research
A research synthesis, against strong baselines.
This planned report will test a Rhizome research synthesis against leading systems on a fixed rubric. The run, report, and comparison artifacts are not published yet.
Coordination trace
The cognitive process, made visible.
This planned report will turn an instrumented Rhizome run into an event stream, field graph, and policy timeline. Those artifacts are not published yet.
Everything labeled here as a published Rhizome result is produced by Rhizome.
Metrics are actual numbers from real runs. Where a visualization is reconstructed or simplified for readability, we say so plainly and label it. We'd rather show a modest, verifiable result than an impressive one you can't check.
The system is in active development. Signal 01 marks where it stands today; the other reports remain roadmap items until their evidence is ready.
Follow the work in public.
We publish completed runs and maintain the build-in-public devlog as the work lands. Follow along or reach out directly.