Buffalo2024 / Rulora

Pluggable controls for reliable AI agents: validated outputs, bounded loops and frozen collective decisions.

kikoncuo / jevfire

JEV-inspired parallel decisions for CUDA LLMs. One context, many decisions. vLLM API, game-agent examples, and reproducible benchmarks.