devops

devops

/devops139

All things DevOps

Small batch size is a core principle of DevOps. If it’s not obvious why, think of a toilet. 🚽💩

Imagine trying to flush a toilet after a dozen people use it without flushing. It wouldn’t be pretty. It’s the same for code changes. The smaller and faster you can get them into production, the cleaner and safer they’ll be.

https://open.substack.com/pub/someofthethings/p/toilets-shadows-and-tip-jars?r=jgz3m&utm_campaign=post&utm_medium=web

just dropped a new article on integrating Open WebUI with DataDog LLM Observability, hope you like it 🪄

https://medium.com/@0xthresh/monitor-open-webui-with-datadog-llm-observability-620ef3a598c6

The best piece of info I got from DevOps Days Seattle yesterday was the concept of the LORM stack, coined by John Willis, which attempts to break down the common Gen-AI components in a stack that’s easier to understand. He pulled inspiration from the LAMP stack, which has defined the primary web stack for years.

if anyone will be at DevOps Days Seattle, I’d love to connect!

today’s workday highlight: I un-fucked a CI/CD pipeline I didn’t write in 30 minutes 🔥

Been doing some of the most interesting cross-team alignment work I’ve ever done to bring three startups together under the same enterprise, including taking great modules and practices from all of them to build out our new DevOps standards. It helps having smart devs who are open to collaboration!

Anyone have any favorite go-to tools/ sites for doing cloud pricing comparisons? I know what I'm doing across the various clouds to do it manually, but would love to save some time if I can ☁️

Getting ready to take my CKA exam in two weeks and haven’t been studying at all, wish me luck 🥲

The fact that Fluence mainnet and Arweave AO both drop on the same day next week is KILLING ME, I've been waiting to play with decentralized compute for as long as I've been in web3

gm, this is how I’m spending my Sunday

DevOps debate of the day: where should monitoring/ alerting live in IaC? Should it live close to the component being monitored (for example, defining a DataDog alert in the same folder as the AWS service it’s monitoring) or should monitoring definitions be placed somewhere separate so it can be managed together?

I met one of the founding engineers of this startup today and was introduced to the concept of Causal AI.

Are you guys familiar with causal AI? It was new to me…
will drop a few more notes below for more context

https://www.causely.io

gm, happy No Change Friday

live view of me running a Terraform destroy on the EKS cluster that's been the bane of my existence

coming from Terraform and HCL world, Pulumi has been a bit tough to get my head around. I finally just got it working with a test use case I’m excited about though 🚀

today's project has been spinning up a Pocket node and writing my own script to save myself from future copy-and-paste from Pocket docs, hope this helps someone else: https://github.com/0xThresh/pokt-node-setup/

gm, looking forward to casting about DevOps with you all