Arize AI

Jev-as-a-Judge is live in Arize AX.

Ask most LLM judges a yes/no question and you get a paragraph back. Then you parse it to figure out the answer.

Jev returns typed outputs instead: booleans, choices, or rubric scores, with confidence on classifications. You can also evaluate multiple questions in a single call.

Docs --> arize.com/docs/ax/evalua…

Qwen3.8 Flash is 30x to 40x cheaper per token than Opus 5.5. In a manager/worker loop on 30 hard coding problems, it solved 23 to Opus's 24, cost 15% less per solve, and took a median 37 minutes instead of 21 seconds.

How to test a model switch yourself: arize.com/blog/how-to-te…

As @PepsiCo scales GenAI across its digital platforms, Charles Holive points to visibility, control, and insights as essential for building systems the team can trust.

Thanks to Charles and the PepsiCo team for sharing their perspective! 👏

Inside Phoenix MCP: remote MCP, SQL, and code mode let agents query traces x.com/i/broadcasts/2…

Data built for reporting isn't automatically ready for autonomous AI agents.

Ed Haruni at Air Canada breaks down four levels of data readiness: reporting, prediction, assistance, and autonomous action.

Read Ed's article 👇
linkedin.com/pulse/from-ai-…