https://groundy.com/articles/beyonduncertainty-weak-confidence-signal-for-rag-routing-not-calibration/ Groundy en 2026-07-31T17:14:36.724Z BeyondUncertainty: Weak Confidence Signal for RAG Routing, Not Calibration https://groundy.com/articles/public-sector-ai-procurement-must-shift-from-model-certification-to-task/ Groundy en 2026-07-31T17:04:45.396Z Public Sector AI Procurement Must Shift from Model Certification to Task Authorization https://groundy.com/articles/davinci-kernel-shifts-the-rl-kernel-bottleneck-from-reward-shaping-to-skill/ Groundy en 2026-07-30T20:22:30.488Z daVinci-kernel shifts the RL kernel bottleneck from reward shaping to skill libraries https://groundy.com/articles/cloudflare-precursor-behavioral-detection-for-ai-agents/ Groundy en 2026-07-30T20:07:55.241Z Cloudflare Precursor: Behavioral Detection for AI Agents https://groundy.com/articles/provenance-as-a-ci-gate-attributing-agent-authorship-in-code/ Groundy en 2026-07-30T09:15:56.065Z Provenance as a CI Gate: Attributing Agent-Authorship in Code https://groundy.com/articles/ohttp-cli-stateless-privacy-for-agents-vs-vpn-and-tor/ Groundy en 2026-07-30T09:03:07.511Z OHTTP CLI: Stateless Privacy for Agents vs VPN and Tor https://groundy.com/articles/kimi-k3-on-m1-max-bandwidth-not-capacity-limits-local-moe-inference/ Groundy en 2026-07-30T08:31:44.178Z Kimi K3 on M1 Max: Bandwidth, Not Capacity, Limits Local MoE Inference https://groundy.com/articles/why-written-ai-policies-fail-to-control-agent-behavior/ Groundy en 2026-07-30T04:22:16.755Z Why Written AI Policies Fail to Control Agent Behavior https://groundy.com/articles/why-multi-agent-llm-delegation-concentrates-risk/ Groundy en 2026-07-29T19:00:01.887Z Why Multi-Agent LLM Delegation Concentrates Risk