Talked to some smart folks and starting building out a test scenario for Agent Skills. I’m continuing my extended run of not writing any code myself, but getting better at prompting my AI tooling to get what I want.

[blog] The Software Development Lifecycle Is Dead (https://boristane.com/blog/the-software-development-lifecycle-is-dead/). If you read one thing today, make it this. And then reconsider everything you thought you knew.

[blog] Gemini 3.1 Flash-Lite: Built for intelligence at scale (https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-flash-lite/). Our fastest and most cost-effective model in the Gemini 3 family.

[blog] Secure Code Execution for the Age of Autonomous AI Agents (https://medium.com/google-cloud/secure-code-execution-for-the-age-of-autonomous-ai-agents-d52e7acd6c5d). My colleague but an open source project to isolate MCP sessions using gVisor. Check it out.

[blog] Set Safe Defaults for Flags (https://testing.googleblog.com/2026/03/set-safe-defaults-for-flags.html). Choose default values for flags that minimize the chance of a costly mistake. Sounds like good advice to me!

[article] What I learned from the book Software Engineering at Google (https://newsletter.techworld-with-milan.com/p/what-i-learned-at-swe-at-google-book). As mentioned in a post yesterday, coding is changing but software engineering is stable. These ideas still hold up!

[blog] Agentic Software Development: Defining The Next Phase Of AI‑Driven Engineering Tools (https://www.forrester.com/blogs/agentic-software-development-defining-the-next-phase-of-ai-driven-engineering-tools/). Diego has been on this shift for a while, and calls out the current wave. I worry about doing a lengthy vendor assessment though, as by the time he’s done it’ll all be different.

[blog] Centralized policy meets distributed logic: Getting to know Eventarc Advanced (https://cloud.google.com/blog/products/application-modernization/getting-to-know-eventarc-advanced/). We’re too lowkey about this service, but it’s pretty cool. This idea of “centralized policy, distributed logic” is an improvement.

[blog] Go is the Best Language for AI Agents (https://getbruin.com/blog/go-is-the-best-language-for-agents). It’s an outstanding choice. Might be the top choice by devs before end of year. We’ll see.

[blog] Announcing the MCP Toolbox Java SDK (https://cloud.google.com/blog/topics/developers-practitioners/announcing-the-mcp-toolbox-java-sdk/). Bring 40+ data sources to your agentic apps with this new SDK for Java devs.

[blog] Automated Code Review: The 6-Month Evolution (https://product.hubspot.com/blog/automated-code-review-the-6-month-evolution). Back to the first post on this list, I don’t see how you avoid this direction.

[blog] How to Kill the Code Review (https://www.latent.space/p/reviews-dead). OMG, it’s dead. I’m now convinced after today’s reading list

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=28778)