LLM agent evaluation before you ship to production
Offline evaluation for LLM agents: routing accuracy, LLM-as-judge scoring, and RAG faithfulness, plus the quality gate I use to block a deploy when the scores come in under threshold.
Offline evaluation for LLM agents: routing accuracy, LLM-as-judge scoring, and RAG faithfulness, plus the quality gate I use to block a deploy when the scores come in under threshold.
WordPress.org now runs an MCP server for the Plugin Directory. It lets AI editors like Cursor and Claude validate readme.txt, check submission status, and submit plugins without leaving the editor.
AI 0.6.0 turns experiments into a type of feature, adds iterative image editing in the Media Library, and renames hooks. C2PA content provenance is queued for 0.7.0.
How to linearize a nonlinear objective with piecewise linear approximations, keep the segments adjacent using SOS type 2, and hand the result to an ordinary LP/MIP solver like Gurobi.
The CSS corner-shape property paired with a scroll timeline, using superellipse() exponents instead of keywords, plus the isolation fix for mix-blend-mode breaking your stacking context.
WordPress.com has added 19 write abilities to its MCP integration, so an agent can create posts, pages, media and terms. Drafts by default, explicit approval before every change, and your own role still caps what it can reach.
WordPress 7.0 gives paragraphs native indentation with no custom CSS. How to opt a custom block in from block.json, choose between the subsequent and all selectors, and set the default in theme.json.
Most agencies build their own stack and end up debugging Nginx instead of shipping. What plugin bloat really costs, and what Encircle Technologies changed when they moved to managed hosting.
Scott Wilson came out of a liver transplant with an organ donation campaign to run, not time to debug a slider. Why he had his site built for him instead of assembling it himself.
A readmission model with 94% accuracy made things worse because it found a correlation, not a cause. Where Pearl’s ladder of causation comes in, and how DoWhy’s four steps keep you honest.