I’ve been home for a few weeks, which has been nice. Next week starts a run of four trips in four weeks. Next week, I’ll be up in Sunnyvale for an Agentic AI event. GitHub Universe the week after, Google corporate stuff after that, then a week in Japan.
[blog] How AI has changed my engineering leadership role. This should register with you if you’re a manager. If not, it might be worth asking why.
[blog] state of agentic coding #11: the curiosity test, lisp, and labs that can’t fail. Hard truths, feels accurate. You have access to the same tools, apps, and hours-in-the-day that the world’ best innovators do. What are you doing with it?
[blog] Context Inertia: When Agents Won’t Let Go. Does your agent or LLM act like a dog with a bone? Here’s a path, and it won’t deviate? Drew gives that a name, and talks about ways to mitigate. One way is to simply watch what’s going on and step in.
[blog] Setting Up an Antigravity Workspace for Multiple Conversations. I like seeing how people personalize their own setup. You may not have the same use case or style as Mete, and that’s ok. We’ll tailor our environment to what works for each of us.
[blog] Day 0 support for Reflection’s Beam on Google Cloud: 5x faster perf with llm-d and NVIDIA Dynamo. Super impressive model (see this great post about it from HuggingFace) and we’ve done some work with them to optimize it on Google Cloud.
[article] Trends #12: We ship 3x more code and review it the same way. Good roundup of recent trends, with useful commentary.
[blog] Evaluating AI agents: moving past the vibe check. Assemble something or buy something, but figure out how to invest more into building evals for your agents.
[blog] Interactive refactoring in gopls. Pretty slick. Human developers will love this, and I can imagine this being useful to coding agents in some way too.
[blog] Welcome Muse Spark Open: open-weight frontier multimodal mixture-of-experts model from Meta. The Hugging Face blogs are always better than the vendor ones. Here they dig into this impressive open-weight US model.
[blog] Bringing the magic: Google breaks open a bottleneck in quantum computing. Whether it’s flying cars or quantum computing, I’m impressed by things I can barely understand.
[youtube-video] Every app on Earth is now open source… whether it likes it or not. We can reverse engineering most any software now? Um, what does that mean? Probably a lot.
[blog] More Agents, Same Mythical Month. Your agent teams may look suspiciously like your existing org structure. Here are some things to consider.
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:
Leave a Reply