Content

AI, XR, DevOps, graph theory, knowledge graphs, and digital sovereignty. Deep dives, tutorials, and analysis from across the GraphWiz knowledge base.

Mapping and Porting Software with CodeGraphContext and Agentflow: The World-Office Case

Case study: CGC and Agentflow produced a ten-task port plan mapping an OnlyOffice-derived C++ engine to Rust/WASM for World-Office.

codegraphcontextagentflowsoftware-portingknowledge-graphs+6 more

November 8, 2026· ~12 min read

Does 85% Confidence Mean 85% Correct? Calibrating a Decision Model on a 12 GB Laptop GPU

A hands-on evaluation of a local Jev-style decision pipeline on an RTX 4080 Laptop: raw logits are overconfident by 13 points, a fitted temperature fixes the cliff but not everything, and one blind self-repair round rescues 40% of the misses — no answer key in the prompt.

jev

calibration

llama-cpp

decision-model

+5 more

November 6, 2026· ~8 min read

Four Billion Parameters of Three.js: Benchmarking a 4B 3D-Coder's Raw Output and Its Self-Repair Loop

A hands-on evaluation of Qwen3.5-4B-3D-Coder with MTP speculative decoding on a 12 GB laptop GPU: 0 of 3 generated scenes run unpatched, yet the model debugs all of them in one self-repair round.

qwenthree.jsllmspeculative-decoding+5 more

November 2, 2026· ~5 min read

Time-Travel Retrieval for Agent Memory: Temporal Knowledge Graphs in Practice

How temporal knowledge graphs give AI agents durable, auditable memory — and why time-travel retrieval is harder than it looks.

agent-memorytemporal-knowledge-graphsgraphragneo4j+1 more

November 1, 2026· ~7 min read

White Font, Real Fines: A Brazilian Court Treats Prompt Injection as an Attack on Justice

Two law firms hid an instruction to a court's AI in white text on a white background. The 3rd Labour Court of Parauapebas fined them 10% of the claim — apparently the first ruling to treat a court filing's machine-readable layer as a legal surface.

prompt-injectionai-securityragdocument-pipelines+2 more

October 26, 2026· ~9 min read

One Discarded Return Value: Auditing Broccoli2Shell Across a Live Fleet

One swallowed error code in the Brotli decoder became a critical RCE spanning Node.js, Bun, .NET and Python. An eight-host audit shows why version scans mislead.

cve-2026-111497brotlisupply-chain-securitynodejs+2 more