Daily Reading List – December 19, 2023 (#227)

As I complete tasks in front of me, I’m hoping to have some cycles on Wednesday or Thursday to experiment with new tech and build something. That feeling of seeing an app compile and start up is still a rush!

[blog] VideoPoet: A large language model for zero-shot video generation. AI-driven video generation is wild. Especially using just a single LLM. Check out this post for some pretty amazing example videos. And skim through the standalone site for the project.

[blog] Ray on Vertex AI: Let’s get it started. Plenty of folks run frameworks like Ray on VMs or Kubernetes clusters. But if you want to run and scale AI and Python apps without worrying about infrastructure, this is a post worth reading.

[blog] Are the Big 4 about to get their lunch eaten by GenAI? Is generative “just another technology tool” or a disruptor the world’s biggest IT consulting firms? Probably the latter.

[blog] More than 10k scientific papers were retracted in 2023. Even the “experts” make mistakes, and a healthy disregard for “consensus” is probably not a terrible thing.

[docs] Build hybrid and multicloud architectures using Google Cloud. What passes as “guidance” about hybrid and multicloud strategy in our industry is honestly pretty weak. You see some generic tips, but very little that’s actionable. This new three-part series in our docs is 140 pages of the best content I’ve seen on the topic.

[article] Practical Magic: Improving Productivity and Happiness for Software Development Teams. The LinkedIn Engineering team open sourced their framework for understanding developers and their needs. Good resource!

[blog] Migrating from Cassandra to Bigtable at Latin America’s largest streaming service. A lift-and-shift migration to the cloud(s)—and maybe swapping a self-managed version of an infrastructure component for a managed version—is probably the safest bet. It’s also the least satisfying for most as they wonder why the performance, scale, and cost don’t match up with their expectations. This is a good story of a replatforming that paid off.

[blog] Product Predictions 2024. Spicy wisdom and opinions from Marty in this post. Product management might be in the “trough of disillusionment” right now for many leaders, but strong product leadership has never been more important.

[blog] Democratizing access to AI-enabled coding with Colab. There aren’t many places with more monthly active devs than Colab. Now those 10+ million folks have access to even more AI assistance.

##

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