This was one of those weekends that makes the high California taxes feel entirely worth it. 70 degrees and sunny in January; I’ll never get used to it. I haven’t seen much sun today since I’ve been locked in my office all day, but I did read a lot, so that’s a plus.
[blog] 10 Benefits and 10 Challenges of Applying Large Language Models to DoD Software Acquisition. If you’re in IT, you probably dread the RFI/RFP process of selecting software products. This post offers an interesting use case for generative AI making that process easier for humans.
[paper] Lumiere: A Space-Time Diffusion Model for Video Generation. Is this the opposite of using generative AI for software acquisition? Probably. This is about realistic motion from text to video production. Read this paper, or check out the project’s site and video.
[paper] DevEx in Action: A study of its tangible impacts. Smart folks are looking at developer experience and creating frameworks for understanding and measuring it. Read this new paper, and the summary blog by one of the authors.
[blog] How to build a gen AI application. Do you know what a “generative AI app” is? This post explains some of the design considerations we’ve learned.
[blog] Pixel guessing : using Gemini Pro Vision with Go. Can generative AI parse pixel-y pictures and figure out what the underlying image actually is? Val has a fun demo that you can try out yourself online.
[blog] Rust vs Go in 2024. Use whatever mainstream programming language makes you happy. There are very few bad choices you can make!
[blog] Spanner integration testing with the emulator. Software emulators make local development a little easier, and helps with integration testing too. Here’s how to use test our amazing database locally in your GitHub Actions.
[article] The One Billion Row Challenge Shows That Java Can Process a One Billion Rows File in Two Seconds. Fun stuff. Gunnar kicked up the “One Billion Row Challenge” and folks are showing creative ways to read 1B records in a couple of seconds.
[article] Google extends its Project IDX development environment with built-in iOS and Android emulators. This experiment from Google offers a cloud-based IDE that targets frontend, mobile, and backend devs. This is a cool update for mobile devs. More here from InfoWorld.
[blog] How we built Project IDX. Lots of transparency in this post by the team building Project IDX. They talk about their architecture, tech choices, and prioritized use cases.
[article] Why IT execs need to consider GraphQL. Should executives care about implementation tech or specs? I dunno. But there’s something to what GraphQL offers.
##
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:
Richard Seroter's Architecture Musings 
One thought