Today I came across some solid technical content and a handful of how-tos that you might be interested in. Take a look!
[blog] Letting our feature flags fly. Nice post from the McDonald’s Engineering team where they explain their need for feature flags in their mobile app, and their architecture for supporting it.
[repo] grunner – Self hosted GitHub Actions runners on GCP using GCE. GitHub Actions is a popular CI solution, and you can host the main workers outside of GitHub. Here’s an example of hosting these runners on Google Cloud.
[article] The 5 most essential technologies for enterprises today. There’s no single list of what you should care about, as it’s contextual to your own business. But this list of items represents a reasonable set of concerns for any enterprise.
[blog] Using Google Cloud from Colab. Millions of devs use Colab to host Jupyter notebooks and experiment with Python. It’s also not too difficult to connect your notebook to Google Cloud to access a range of other services.
[blog] Generative AI Learning Path Notes — Part 1. Why take the training yourself if you can just glance at the notes of a more studious colleague? Mete published his notes for generative AI coursework, and there’s some good knowledge to pick up.
[blog] gRPC Service to Service on Cloud Run. I admittedly default to basic REST/HTTP requests when I build components that talk to each other, but really should try using gRPC more often. This post goes into some depth on setting that up.
[article] Discord Migrates Trillions of Messages from Cassandra to ScyllaDB. Database migrations are no joke, and Discord doing a couple of major ones over these past few years is interesting to me.
[blog] Go 1.21 Release Candidate. There are some good language fundamentals showing up here, as well as some new Web Assembly support.
[blog] Analyzing Volatile Memory on a Google Kubernetes Engine Node. The Spotify team use a lot of GKE in production (hundreds of thousands of pods), but how do they look for malicious behavior on a given GKE node? Here’s their approach.
[article] Treat Your CI System as a Product for Faster and Better Feedback. Dev teams spend a LOT of time waiting for builds and tests to complete. Investing in your CI platform and treating it like a product can have major rewards.
##
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