Do ever have those “perform research” days where you know your brain will be running a background thread even after you’re done working? I can sense it, after a day of investigating a handful of distinct areas.
[article] How AI will shape software engineering in 2026. Solid piece that covers a lot of areas I see teams navigating right now.
[article] How to hire a chief of staff. Such an important role! My CoS is invaluable and changed how we work.
[blog] Thomas Kurian Explains the Discipline Behind Google Cloud’s Growth. The boss does an excellent job explaining our customer focus and how that drives our strategy. More here.
[blog] Small projects, clear scope. Good, quick advice on the importance of planning and delivering in small batches.
[article] Banks aim for agentic AI scale in 2026: report. It seems like AI is getting embedded into key functions fairly quickly.
[blog] Introducing BigQuery managed and SQL-native inference for open models. This is fantastic. Now use *any* open model within BigQuery for embeddings and inference, while getting automatic resource management.
[blog] AI’s Atmospheric Reentry Begins. Is the free ride over? No, but there’s definitely growing restrictions on unlimited use at virtually no cost.
[article] From typos to takeovers: Inside the industrialization of npm supply chain attacks. Lots of problems called out, no solutions. But there are a handful of ways you can sandbox dependencies, do downstream checks, and other techniques. But do something!
[blog] TranslateGemma: A new suite of open translation models. Now you have fully open translation models for 55 languages, and device-friendly sizes.
[blog] Choosing the Right Multi-Agent Architecture. Handy look at common patterns, why you’d use each one, what tradeoffs you face, and performance implications.
[article] Better Context Will Always Beat a Better Model. That could be true. I’m not sure if a subpar model with exceptional context beats a world-class model with average context.
[blog] Gemini introduces Personal Intelligence. When your AI assistant remembers its history with you, that’s helpful. When it “knows” your overall digital history, it becomes massively useful.
[article] How To Choose the Right Tool for Your Google ADK Agent. There’s more than one type of tool and the one you pick has implications on your architecture.
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:
Richard Seroter's Architecture Musings 
Leave a comment