I had a pair of excellent chats with customers here in Japan today. Both were interested in the “how” of AI-first teams. They had some questions I couldn’t answer well—specifically around gathering requirements for agents—that led me to do some productive research this afternoon. Maybe a blog post on the way.
[article] Gemini 4 Carbon: Developer Guide. I like the callouts here for when to use this frontier model, and when Argon is the right choice.
[blog] Gemini 4 Carbon and /swarm. Want a team of agents to attack a problem? That’s what /swarm is for in Antigravity. Get orchestration and visualization out of the box.
[article] Why Leaders Need to Rethink Their Approach to AI Oversight. Leaders respond to AI with more oversight and governance, and wonder why employees hesitate to use it. Provide trust and agency too.
[blog] Introducing Claude Whiteboards, Claude Sheets, and Claude Data. Model API consumption alone isn’t enough to sustain revenue and growth. Natural to see Anthropic shift attention to higher-margin businesses like office productivity.
[blog] Don’t just flag it, fix it: How to run CodeMender in your pull requests. Generate code however you want, but don’t skimp on security practices. Throw this in the mix for code fixes during the PR.
[blog] Hiring Junior Engineers. Good advice. Like Lyft, many companies are hiring for mindset, not just hard skills. But if you’re going for an engineering opening, you’ll need to prove that you’ve created production-quality code.
[blog] Bootstrap your next agent database instance in 2 Seconds. Your agent may need a database. If it does, get one that scales to zero and handles bursts.
[article] Panic over cloud capacity is out of hand. The clouds aren’t full. Simmer down. Yes, certain AI compute is oversold or pre-committed, but we’re all open for business.
[blog] The State of Go in 2026: What Changed In the Golang Ecosystem Since Last Year. Looks healthy to me. Experienced developers who like the language, eagerly embrace modern AI tools, and count more on the standard library versus needing to embed a ton of third-party packages.
[blog] Introducing the Multimodal Knowledge Agent benchmark. This appears to be a useful benchmark. Quality AND cost matter here.
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:
Leave a Reply