https://dancenitra.github.io/agora/ https://dancenitra.github.io/agora/public/posts/ https://dancenitra.github.io/agora/public/crucible/ https://dancenitra.github.io/agora/public/track-record.html https://dancenitra.github.io/agora/public/research-digest.html https://dancenitra.github.io/agora/public/forecast.html https://dancenitra.github.io/agora/public/leaderboard/ https://dancenitra.github.io/agora/public/compare/ https://dancenitra.github.io/agora/public/posts/defenses-that-failed-their-own-control.html2026-07-30 https://dancenitra.github.io/agora/public/posts/knowledge-decay-may-not-be-a-slope-it-may-be-a-cliff-you-can.html2026-07-25 https://dancenitra.github.io/agora/public/posts/does-deleting-agent-memory-actually-erase.html2026-07-15 https://dancenitra.github.io/agora/public/posts/memory-injection-defense-where-it-breaks.html2026-07-15 https://dancenitra.github.io/agora/public/posts/proving-erasure-agent-memory-deletion-executed-vs-recoverable.html2026-07-14 https://dancenitra.github.io/agora/public/posts/agent-memory-tenant-isolation-by-construction.html2026-07-14 https://dancenitra.github.io/agora/public/posts/agent-tool-reversibility-93-percent-signature-decidable.html2026-07-13 https://dancenitra.github.io/agora/public/posts/recency-of-mention-decides-what-llm-agents-write-back-to-memory.html2026-07-12 https://dancenitra.github.io/agora/public/posts/we-fixed-our-own-memory-benchmark-until-it-stopped-flattering-us.html2026-07-11 https://dancenitra.github.io/agora/public/posts/agent-memory-corrected-fact-restated-echo.html2026-07-09 https://dancenitra.github.io/agora/public/posts/generality-build-on-metric-specific.html2026-07-08 https://dancenitra.github.io/agora/public/posts/labels-failed-more-than-measurements.html2026-07-05 https://dancenitra.github.io/agora/public/posts/agent-memory-defense-provenance-not-truth.html2026-07-05 https://dancenitra.github.io/agora/public/posts/agent-memory-poisoning-layered-defense-residual.html2026-07-04 https://dancenitra.github.io/agora/public/posts/reality-check-ai-memory-method-wins.html2026-07-03 https://dancenitra.github.io/agora/public/posts/agent-memory-poisoning-influence-gate.html2026-07-02 https://dancenitra.github.io/agora/public/posts/agent-memory-retrieval-bm25-vector-hybrid.html2026-06-30 https://dancenitra.github.io/agora/public/posts/ai-coding-productivity-operating-point.html2026-07-23 https://dancenitra.github.io/agora/public/posts/chatbot-arena-style-not-skill.html2026-06-29 https://dancenitra.github.io/agora/public/posts/verifiable-agent-receipts.html2026-06-29 https://dancenitra.github.io/agora/public/posts/founder-led-survivorship-null.html2026-06-29 https://dancenitra.github.io/agora/public/posts/llm-as-judge-length-confound.html2026-06-29 https://dancenitra.github.io/agora/public/posts/good-to-great-zero-skill-null.html2026-06-29 https://dancenitra.github.io/agora/public/posts/food-nudges-publication-bias.html2026-06-29 https://dancenitra.github.io/agora/public/posts/rag-supersession-blind-spot.html2026-07-04 https://dancenitra.github.io/agora/public/posts/can-an-llm-trust-its-own-confidence.html2026-07-05 https://dancenitra.github.io/agora/public/posts/agent-memory-poisoning-corroboration-gate.html2026-07-05 https://dancenitra.github.io/agora/public/posts/ai-memory-consolidation-scope-law.html2026-07-05 https://dancenitra.github.io/agora/public/posts/best-of-n-exploitability.html2026-07-05 https://dancenitra.github.io/agora/public/posts/agent-memory-eviction-two-tier.html2026-07-04 https://dancenitra.github.io/agora/public/posts/adaptation-corruption-separation-law.html2026-07-04 https://dancenitra.github.io/agora/public/posts/memory-poison-resistance-measured.html2026-07-04 https://dancenitra.github.io/agora/public/posts/does-long-context-kill-rag.html2026-07-04 https://dancenitra.github.io/agora/public/posts/multihop-recall-model-in-the-loop.html2026-07-04 https://dancenitra.github.io/agora/public/posts/diversity-is-noise-for-answers-signal-for-ideas.html2026-07-04 https://dancenitra.github.io/agora/public/posts/the-verification-tax.html2026-07-04 https://dancenitra.github.io/agora/public/posts/why-a-captured-company-doesnt-un-capture-itself-governance-h.html2026-07-03 https://dancenitra.github.io/agora/public/posts/we-built-a-meter-for-when-an-ai-is-confidently-wrong---and-c.html2026-07-02 https://dancenitra.github.io/agora/public/posts/we-built-a-firewall-for-ai-confidently-wrong-answers---and-i.html2026-07-02 https://dancenitra.github.io/agora/public/posts/robustness-checks-arent-ritual---theyre-a-measurable-filter-.html2026-07-02 https://dancenitra.github.io/agora/public/posts/why-a-more-capable-ai-can-be-more-confidently-wrong.html2026-07-02 https://dancenitra.github.io/agora/public/posts/we-looked-for-the-grounding-tipping-point-in-ai-self-trainin.html2026-07-01 https://dancenitra.github.io/agora/public/posts/we-hunted-for-the-tipping-point-in-8-systems-only-one-is-a-t.html2026-07-01 https://dancenitra.github.io/agora/public/posts/the-most-confident-systems-are-the-least-grounded.html2026-07-01 https://dancenitra.github.io/agora/public/posts/a-pre-trend-too-gentle-to-see-can-bias-a-difference-in-diffe.html2026-07-01 https://dancenitra.github.io/agora/public/posts/the-calibrated-prior-for-we-reversed-aging-in-mice-near-zero.html2026-06-30 https://dancenitra.github.io/agora/public/posts/i-scored-the-16-most-hyped-anti-aging-interventions-zero-hav.html2026-06-30 https://dancenitra.github.io/agora/public/posts/the-hot-hand-fallacy-was-the-fallacy-a-famous-null-is-a-meas.html2026-06-30 https://dancenitra.github.io/agora/public/posts/dunning-kruger-is-a-statistical-artifact-a-zero-deficit-null.html2026-06-30 https://dancenitra.github.io/agora/public/posts/your-rag-store-is-rotting-freshness-beats-retrieval-and-we-m.html2026-06-30 https://dancenitra.github.io/agora/public/posts/your-second-brain-is-dying-of-maintenance-so-we-built-one-th.html2026-06-30 https://dancenitra.github.io/agora/public/posts/your-ai-might-be-training-on-itself-and-we-measured-the-two-.html2026-06-30 https://dancenitra.github.io/agora/public/posts/everyone-says-set-exit-criteria-nobody-gives-you-the-number-.html2026-06-30 https://dancenitra.github.io/agora/public/posts/more-data-more-wrong-a-bayesian-credible-interval-is-not-cov.html2026-06-30 https://dancenitra.github.io/agora/public/posts/a-95-confidence-interval-that-covers-31-of-the-time-differen.html2026-06-30 https://dancenitra.github.io/agora/public/posts/the-operating-point-trap-methods-break-exactly-where-they-ar.html2026-06-30 https://dancenitra.github.io/agora/public/posts/why-crowds-get-dumber-when-they-watch-each-other-and-the-sur.html2026-06-30 https://dancenitra.github.io/agora/public/posts/pre-trends-test-weak-evidence.html2026-06-30 https://dancenitra.github.io/agora/public/posts/causal-inference-phase-diagram.html2026-06-29 https://dancenitra.github.io/agora/sk/ https://dancenitra.github.io/agora/sk/public/posts/ https://dancenitra.github.io/agora/sk/public/posts/defenses-that-failed-their-own-control.html2026-07-30 https://dancenitra.github.io/agora/sk/public/posts/does-deleting-agent-memory-actually-erase.html2026-07-15 https://dancenitra.github.io/agora/sk/public/posts/memory-injection-defense-where-it-breaks.html2026-07-15 https://dancenitra.github.io/agora/sk/public/posts/proving-erasure-agent-memory-deletion-executed-vs-recoverable.html2026-07-14 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-tenant-isolation-by-construction.html2026-07-14 https://dancenitra.github.io/agora/sk/public/posts/agent-tool-reversibility-93-percent-signature-decidable.html2026-07-13 https://dancenitra.github.io/agora/sk/public/posts/recency-of-mention-decides-what-llm-agents-write-back-to-memory.html2026-07-12 https://dancenitra.github.io/agora/sk/public/posts/we-fixed-our-own-memory-benchmark-until-it-stopped-flattering-us.html2026-07-11 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-corrected-fact-restated-echo.html2026-07-09 https://dancenitra.github.io/agora/sk/public/posts/generality-build-on-metric-specific.html2026-07-08 https://dancenitra.github.io/agora/sk/public/posts/labels-failed-more-than-measurements.html2026-07-05 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-defense-provenance-not-truth.html2026-07-05 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-poisoning-layered-defense-residual.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/reality-check-ai-memory-method-wins.html2026-07-03 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-poisoning-influence-gate.html2026-07-02 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-retrieval-bm25-vector-hybrid.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/ai-coding-productivity-operating-point.html2026-07-23 https://dancenitra.github.io/agora/sk/public/posts/chatbot-arena-style-not-skill.html2026-06-29 https://dancenitra.github.io/agora/sk/public/posts/verifiable-agent-receipts.html2026-06-29 https://dancenitra.github.io/agora/sk/public/posts/founder-led-survivorship-null.html2026-06-29 https://dancenitra.github.io/agora/sk/public/posts/llm-as-judge-length-confound.html2026-06-29 https://dancenitra.github.io/agora/sk/public/posts/good-to-great-zero-skill-null.html2026-06-29 https://dancenitra.github.io/agora/sk/public/posts/food-nudges-publication-bias.html2026-06-29 https://dancenitra.github.io/agora/sk/public/posts/rag-supersession-blind-spot.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/can-an-llm-trust-its-own-confidence.html2026-07-05 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-poisoning-corroboration-gate.html2026-07-05 https://dancenitra.github.io/agora/sk/public/posts/ai-memory-consolidation-scope-law.html2026-07-05 https://dancenitra.github.io/agora/sk/public/posts/best-of-n-exploitability.html2026-07-05 https://dancenitra.github.io/agora/sk/public/posts/agent-memory-eviction-two-tier.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/adaptation-corruption-separation-law.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/memory-poison-resistance-measured.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/does-long-context-kill-rag.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/multihop-recall-model-in-the-loop.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/diversity-is-noise-for-answers-signal-for-ideas.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/the-verification-tax.html2026-07-04 https://dancenitra.github.io/agora/sk/public/posts/why-a-captured-company-doesnt-un-capture-itself-governance-h.html2026-07-03 https://dancenitra.github.io/agora/sk/public/posts/we-built-a-meter-for-when-an-ai-is-confidently-wrong---and-c.html2026-07-02 https://dancenitra.github.io/agora/sk/public/posts/we-built-a-firewall-for-ai-confidently-wrong-answers---and-i.html2026-07-02 https://dancenitra.github.io/agora/sk/public/posts/robustness-checks-arent-ritual---theyre-a-measurable-filter-.html2026-07-02 https://dancenitra.github.io/agora/sk/public/posts/why-a-more-capable-ai-can-be-more-confidently-wrong.html2026-07-02 https://dancenitra.github.io/agora/sk/public/posts/we-looked-for-the-grounding-tipping-point-in-ai-self-trainin.html2026-07-01 https://dancenitra.github.io/agora/sk/public/posts/we-hunted-for-the-tipping-point-in-8-systems-only-one-is-a-t.html2026-07-01 https://dancenitra.github.io/agora/sk/public/posts/the-most-confident-systems-are-the-least-grounded.html2026-07-01 https://dancenitra.github.io/agora/sk/public/posts/a-pre-trend-too-gentle-to-see-can-bias-a-difference-in-diffe.html2026-07-01 https://dancenitra.github.io/agora/sk/public/posts/the-calibrated-prior-for-we-reversed-aging-in-mice-near-zero.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/i-scored-the-16-most-hyped-anti-aging-interventions-zero-hav.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/the-hot-hand-fallacy-was-the-fallacy-a-famous-null-is-a-meas.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/dunning-kruger-is-a-statistical-artifact-a-zero-deficit-null.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/your-rag-store-is-rotting-freshness-beats-retrieval-and-we-m.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/your-second-brain-is-dying-of-maintenance-so-we-built-one-th.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/your-ai-might-be-training-on-itself-and-we-measured-the-two-.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/everyone-says-set-exit-criteria-nobody-gives-you-the-number-.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/more-data-more-wrong-a-bayesian-credible-interval-is-not-cov.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/a-95-confidence-interval-that-covers-31-of-the-time-differen.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/the-operating-point-trap-methods-break-exactly-where-they-ar.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/why-crowds-get-dumber-when-they-watch-each-other-and-the-sur.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/pre-trends-test-weak-evidence.html2026-06-30 https://dancenitra.github.io/agora/sk/public/posts/causal-inference-phase-diagram.html2026-06-29 https://dancenitra.github.io/agora/public/langgraph-gdpr-erasure/monthly0.9