Daily Reading List – May 9, 2025 (#549)
Finishing the week strong with some writing, team work, and a few minutes to try out one of our upcoming products. Enjoy your weekend!
[blog] Advancing the frontier of video understanding with Gemini 2.5. This feels like a game-changer. Video gets some first-class treatment now, whether you’re building apps or applying reasoning.
[blog] Measuring Dialogue Intelligibility for Netflix Content. Neat post from Netflix Engineering about how they detect low or unclear audio and look at ways to improve it for viewers.
[article] When You’re Asked to Meet Impossible Goals. What if you’re the one creating the impossible goals for your team? Asking for a friend.
[blog] Build, use and share data with data products in BigQuery. Here’s a useful way to turn your data assets into a sharable product.
[article] InfoQ Culture and Methods Trends Report – 2025. Here’s one look at what’s mainstream versus emerging around tech culture. This definitely has some changes from last year.
[blog] New column-granularity indexing in BigQuery offers a leap in query performance. You might not think we could make dramatic gains in BigQuery performance at this point. You’d be wrong.
[blog] MCP: What It Is and Why It Matters—Part 1. Hopefully as you’ve read the various MCP posts I’ve shared, the idea is starting to click into place!
[article] AI-Generated Code Needs Refactoring, Say 76% of Developers. Some new data here. Of course, I’d argue that “human-generated code” needs refactoring too.
[blog] Finding Earth Engine Data Doesn’t Have to Be a Struggle: Let AI Agents Do the Heavy Lifting. I like the idea of using agents to execute on a research plan over complex data sets. Here’s a case using Google Earth Engine.
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below: