Technical writing across the Éthiqueia research threads: memory systems, cognitive architectures, agent runtimes, commodity training, fractal cognition.https://yandesbiens.com/enhttps://yandesbiens.com/blog/contextpaw-silent-truncation/https://yandesbiens.com/blog/contextpaw-silent-truncation/Ollama silently drops the front of an oversized prompt and returns 200 OK, so the model confidently invents an answer. llama.cpp is honest and returns a 400 — which kills the agent turn instead. I measured both, then built ContextPaw to make overflow honest and survivable. Includes a result I did not expect: telling an agent it lost information does not stop it hallucinating.Sat, 11 Jul 2026 00:00:00 GMTagentscontextpawollamallama.cppagentscontext-windowbenchmarklocal-firsthttps://yandesbiens.com/blog/hermes-certainty-ratchet/https://yandesbiens.com/blog/hermes-certainty-ratchet/A negative result, measured. An idle 'reasoning' loop ran 27,834 times over months and converged on unfalsifiable certainty: 606 beliefs strengthened per cycle with zero contradictions, 882,602 pieces of self-generated evidence of which none disagreed, and an audit log that recorded 5 of 16.9 million confidence increases. The decay mechanism that should have stopped it was disarmed by the loop's own evidence writes.Sat, 11 Jul 2026 00:00:00 GMTcognitionhermesneuroarchcognitive-architectureautonomous-agentsnegative-resultforensicsepistemicshttps://yandesbiens.com/blog/agentos-agent-platform/https://yandesbiens.com/blog/agentos-agent-platform/How the agentos kernel grew into a self-hosted agent platform: a ReAct loop over 120+ models, ~26 real tools, a permission layer that keeps a human in the loop, runtime tool-forging in a sandbox, and an environment-aware system prompt. Engineering, not a benchmark.Wed, 01 Jul 2026 00:00:00 GMTagentsagentosdevlogagentstool-uselocal-firstsafetyhttps://yandesbiens.com/blog/forgelm-4090-cost/https://yandesbiens.com/blog/forgelm-4090-cost/A measured training envelope (VRAM, throughput, tokens/day) for 30M-500M models on a single 24GB RTX 4090, plus the real 120M loss curve.Mon, 29 Jun 2026 00:00:00 GMTtrainingtrainingbenchmarkhttps://yandesbiens.com/blog/aeon-liquid-brain-stability/https://yandesbiens.com/blog/aeon-liquid-brain-stability/A development log from AEON: how a failing unit test exposed a seed-dependent trainability bug in the CfC liquid network, why a missing normalization was the root cause, and how one LayerNorm made the student learn reliably across every seed.Sun, 28 Jun 2026 00:00:00 GMTcognitionaeondevlogliquid-neural-networkscfctraining-stabilitylocal-firsthttps://yandesbiens.com/blog/ufm-benchmark/https://yandesbiens.com/blog/ufm-benchmark/UFM lets a single RTX 4090 run a routed model whose expert bank doesn't fit in VRAM. Here's the benchmark, the code, and the regime where it doesn't help.Sat, 27 Jun 2026 00:00:00 GMTmemoryufmbenchmarklocal-firstmemorymoertx4090https://yandesbiens.com/blog/fmm-benchmark/https://yandesbiens.com/blog/fmm-benchmark/FMM's topic-scoped retrieval is ~164× faster and ~3.7× more accurate than a flat scan at 128k items — and useless if you misroute the scope. The locality bet, in semantic memory.Sat, 27 Jun 2026 00:00:00 GMTmemoryfmmbenchmarkmemoryretrievalraglocal-firsthttps://yandesbiens.com/blog/fmm-router/https://yandesbiens.com/blog/fmm-router/A near-free centroid router recovers 98% of the oracle's recall at ~60× flat-scan speed — when the memory is separable. As topics overlap, routing (not retrieval) becomes the bottleneck and the win evaporates.Sat, 27 Jun 2026 00:00:00 GMTmemoryfmmbenchmarkmemoryretrievalroutingraglocal-first