W&B adds stronger agent tracing, harness integrations and automated evals in Weave
_ScottCondron · x · 2026-07-23
W&B expands Weave with better agent tracing and eval workflows
The post says the team is improving the UX for tracking agents in Weights & Biases and seeing more teams use the stack for auto-research, production agents, and rollouts during training.
What’s changing
- Broader adoption of GenAI OTEL, which the team says it has adopted wholesale
- Easier integrations with popular harnesses
- More readable agent traces, including markdown-style message and tool-call views
- New ways to inspect traces and improve evals, including:
- prod LLM judges that flag issues automatically
- an integrated agent that queries traces and helps root-cause failures
- better navigation across millions of traces
The overall message is that agent observability is still open territory, and W&B wants to own the tooling layer.
More from coding & agent
- Baseten says it is partnering with Harvey on post-training for long-running agents — baseten · 2026-07-23
- Microsoft open-sources MagenticLite stack, including MagenticBrain and Fara 1.5 — SaleemaAmershi · 2026-07-23
- Thread claims GPT-5.6 Sol helped solve 6 open Erdős problems in 5 days — jxnlco · 2026-07-23
- Gary Marcus says Claude Code uses 50+ tools and looks closer to a neuro-symbolic system — GaryMarcus · 2026-07-23
- Seven questions teams should answer before greenlighting an AI agent pilot — kashifmanzoor · 2026-07-23
- A proxy gateway approach to stopping agent tool-call loops — bulleykebaal · 2026-07-23