As published on wandb.ai, August 2026. Captured by usedby on Oct 3, 2026.
What happened
JetBrains uses W&B Weave as the observability provider for running evaluations of its AI agents, built with its Koog framework. Reviewing the logs helped the team find problems in core AI algorithms.
Summary written by usedby from the source page, in English. The figures are those of Weights & Biases and JetBrains, not ours.
The most obvious outcome that we got from using W&B as our observability provider for running our agent evaluation was figuring out that some of our core AI algorithms that we developed was not working correctly.
The most obvious outcome that we got from using W&B as our observability provider for running our agent evaluation was figuring out that some of our core AI algorithms that we developed was not working correctly. By looking at the logs in the console, we realized there are some issues and we had to rework the whole algorithm.
What the story claims, and what we checked
We compared the story with its live page on Oct 3, 2026.
- The passage quoted aboveCheckedCopied word for word from the page, near the name of JetBrains.
- The publication dateCheckedRead from the page’s own metadata, never guessed.
- JetBrains uses Weights & BiasesCheckedConfirmed line. Latest check across sources: Oct 3, 2026.
- The result itselfNot checkedWe quote it; we did not measure it.




