A multi-agent Discord Senate — four seats, public debate, evidence-first votes, a living constitution under a human Crown, and a Staff account that can never launder executive authority. 253 public motions and counting.
Read more
Adversarial multi-model failure atlas — subscription controls (Claude / Codex / Grok) vs Skynet fleet cells, with flag rates, empty replies, hard errors, and latency.
Read more
Empirical cross-provider LLM benchmark — 22,522 trials × 64 models × 9 tasks, with subscription-vs-public-API cost columns side-by-side and industry-benchmark side tracks.
Read more
A personal AI subscription cost tracker — single view of every AI service I pay for, with spend trends and a browsable plan catalog.
Read more
A calibrated image-generation pipeline talking to SwarmUI on a 3090TI rig — 12 calibration rounds, 260+ graded samples, 18 curated model presets.
Read more