Daily Reading List – May 27, 2025 (#560)
Back after a three-day weekend, and feeling good. A backlog of reading material means longer reading lists this week. You’re getting your money’s worth!
[blog] Reinvent the Wheel. Sure, sometimes you should reinvent the wheel, repeat yourself, and fix what isn’t broken. We often learn and make unexpectedly big improvements when we don’t just accept the status quo.
[blog] Cloud CISO Perspectives: How Google Cloud’s security team helps build securely. Here’s some insight into our engineering mindset, and also a great roundup of links and news.
[article] What causes procrastination for software engineers? Procrastination has both negative AND positive side effects? This research also offers a handful of strategies to mitigate procrastination.
[blog] Journey to 1000 models: Scaling Instagram’s recommendation system. Cool post, and a reminder that adding ML models to your architecture requires a reassessment and redesign of a few dimensions.
[blog] Transforming Kubernetes and GKE into the leading platform for AI/ML. Kubernetes seems well positioned to be a key part of the training and serving infrastructure for companies of any size.
[article] How to Lead an All-Hands After Delivering Bad News. Over the years, I’ve had a few of these. This is excellent advice.
[blog] Vertex AI Studio, redesigned: Your source for generative AI media models across all modalities. Cloud customers deserve nice things. This refresh of the Vertex AI Studio experience makes it much simpler to design prompts, generate media, ground your responses, and more.
[article] What Salesforce’s $8B acquisition of Informatica means for enterprise data and AI. This might have been a bigger headline a few years back, but I still can imagine lots of good value here. More here.
[blog] An ADK Java GitHub template for your first Java AI agent. Sometimes the “getting started demos” can themselves be intimidating. I like this sort of bare bones entry into a new framework. Build on it from here!
[blog] The Triad of Agent Architecture: ADK, MCP, and Cloud Run. Here’s a code-heavy demonstration for an agent solution.
[blog] Google AI Studio: How to go from a prompt to a geo-location guessing app in minutes. Vibe code your way to a running app in no time at all.
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below: