Daily Reading List – May 19, 2025 (#555)

I’m in Silicon Valley today to get ready for Google I/O which starts tomorrow. It’s a welcome distraction as I had a complex weekend. Some great family time, but then one of my two dogs (Ellie) passed away. I loved that pup and had almost thirteen great years of companionship. Dogs are the best.

[blog] Vibe Coding a 5,000km Bike Race: Building a Route Planner with Firebase & Gemini. Cool post from a colleague who had a use case, and quickly solved it with AI assistance.

[blog] Why Are There So Many Databases? This big post looks at nine types of databases and when you’d use each one.

[blog] Understand anything, anywhere with the new NotebookLM app. For both iOS and Android users. Save, listen, and learn on the go.

[paper] AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenge. How do these concepts relate and differ? This paper explores.

[article] Does LLM Write Performant Code? Survey Says No. This assessment showed that LLMs didn’t generate higher performing code very often. Yet?

[blog] Why strong engineers are rarely blocked. There’s a savviness that comes with experience and knowing how to get work done effectively.

[blog] Building Agent2Agent (A2A) Server with Google Apps Script. I’m starting to think you can do almost anything in Google Apps Script. This might be a blueprint for implementing A2A in other unexpected places.

[blog] AI Hypercomputer developer experience enhancements from Q1 25: build faster, scale bigger. ML engineers keep getting better and better tools for training and serving their models.

[article] To Drive Innovation, Create the Conditions for Serendipity. What are the three steps to serendipity that you should understand? Read this to learn.

[article] Google’s AlphaEvolve: The AI agent that reclaimed 0.7% of Google’s compute – and how to copy it. Here’s a deep dive into this system from Google and the benefits we’ve seen as a result of using it.

[article] AWS Launches Its Take on an Open Source AI Agents SDK. Another day, another new framework! I’m sure this one will get some traction given who made it.

[article] OpenAI launches Codex, an AI coding agent, in ChatGPT. AI coding agents are the future, and this new offering from OpenAI is interesting.

[article] Windsurf Launches SWE-1: AI Models Built for the Entire Software Engineering Process. This looks like a cool improvement. I wonder how it evolves now with WIndsurf as part of OpenAI.

Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below: