Mistakes happen. As a team, the important thing is to recognize it’s never an individuals’s fault — it’s the process, the culture, or the infra.
In this case, there was a manual deploy step that should have been better automated. Our team has made a few improvements to the automation for next time, a couple more on the way.
On Claude Code, we’re introducing agent teams.
Spin up multiple agents that coordinate autonomously and work in parallel—best for tasks that can be split up and tackled independently.
Agent teams are in research preview: https://t.co/LdkPjzxFZg
📊 Observability becomes mandatory
With agents calling tools for hours, debugging ≠ stack traces anymore.
It’s about traces, evals, and reasoning failures.
Infra is catching up to reality.
ICYMI: Go beyond standard reports ✨
The open-source MCP server connects your GA data to LLMs like Gemini — making it possible to ask questions, explore insights, and build smarter, data-driven workflows: https://t.co/juCUttBXiN
I'm not joking and this isn't funny. We have been trying to build distributed agent orchestrators at Google since last year. There are various options, not everyone is aligned... I gave Claude Code a description of the problem, it generated what we built last year in an hour.
If you like distributed systems, this paper on distributed databases is worth reading. Although it is based on PostgreSQL, technology should transcend boundaries. https://t.co/3UtwW2llUl
🔧 ReActMCP Integration
Discover a hands-on educational project that combines Mode Context Protocol with LangChain's framework, showcasing powerful integrations of OpenAI and ExaAI's web search capabilities.
Start building with MCP today! 🚀
https://t.co/9htfUHQq3o
💰 Anthropic token-saving updates
Yesterday Anthropic released multiple token-saving updates, including simpler cache management and token-efficient tool use (and a new built-in tool)!
We've updated our documentation to showcase these features, link in 🧵!