About to fly home after an exhausting week in Mountain View. I never realize how much I enjoy that extra monitor at home until I’m stuck with just my laptop screen on trips!

[article] With Android CLI, Google is Making the Android Toolchain Agent-Friendly (https://www.infoq.com/news/2026/05/agent-friendly-android-cli). Android got a lot less intimidating (to me) this week.

[article] AI at scale: What engineering teams are confronting (https://www.infoworld.com/article/4157385/ai-at-scale-what-engineering-teams-are-confronting.html). AI on your machine is one thing. Confidently making it available to others is a different thing.

[blog] What If Lock-In Doesn’t Matter So Much Anymore? (https://kottke.org/26/05/what-if-lock-in-doesnt-matter-so-much-anymore) It’s definitely time to challenge some assumptions. Maybe you have more control now than you thought?

[blog] Agent Sandbox on GKE is now available for everyone, and a first look at Agent Substrate (https://cloud.google.com/blog/products/containers-kubernetes/bringing-you-agent-sandbox-on-gke-and-agent-substrate/). These are both super powerful ways to make Kubernetes a more suitable and dynamic host for agents.

[blog] Prompts are technical debt too (https://www.seangoedecke.com/prompts-are-technical-debt-too/). Yes, these probably degrade faster than most any of your code. Especially if your prompts include explicit references or steps that fell out of date.

[article] Supporting Your Employees’ Career Growth When Everyone Is Overwhelmed (https://hbr.org/2026/05/supporting-your-employees-career-growth-when-everyone-is-overwhelmed). Having a hard time right now with career conversations? It’s not just about promotion discussions, but listening and figuring out a constant learning/development plan.

[blog] Angular Is Exciting Again, and v21 Proves It (https://hackernoon.com/angular-is-exciting-again-and-v21-proves-it). It’s cool when stable, widely-used frameworks keep up with the times.

[blog] The Cost of Safetyism (https://thegrowtheq.com/the-cost-of-safetyism). It’s the parent’s fault. We need to own it. Unsupervised exploration is needed in kids to make them better adults.

[blog] 100 things we announced at I/O 2026 (https://blog.google/innovation-and-ai/technology/ai/google-io-2026-all-our-announcements/). This is a lot, and we didn’t even list a handful of other lowkey items shipped this week.

[blog] What the OSS Summit Says About OSS in 2026 (https://redmonk.com/sogrady/2026/05/19/oss-summit-2026/). What was hot at the Summit? Sounds like open models and maintainer burnout were top of mind.

[blog] Blazing fast on-device GenAI with LiteRT-LM (https://developers.googleblog.com/blazing-fast-on-device-genai-with-litert-lm/). Speedy inference at the edge is opening up a lot of use cases. This framework makes it possible.

[blog] Why Your Coding Agent Can’t Be Your Testing Agent (https://www.devassure.io/blog/why-coding-agents-cant-test). This person suggests you split your author and reviewer to ensure you get a proper asessment.

Want to get this update sent to you every day? Subscribe to my RSS feed (https://seroter.com/feed/) or subscribe via email below:

					Subscribe (https://seroter.com/?post_type=post&p=29960)