Hey {{first name | there}}. This past week I got an email from Netlify saying I was approaching and ultimately had exhausted my quota for deployments. This was strange to me because I did not know such a quota existed.
In today’s technical notes:
Parts of the software stack we often forget about
What I am doing to ensure deployments stay safe
Tools I am liking more and more
📰 TECHNICAL NOTES: The Parts we often forget
When I got the email, I was out on a walk and was quite confused as to how that happened. Then it occurred to me that this is the “velocity” that people often throw around, in the past, there would be a handful of deployments per week, which always kept us under the limits of our plan.
Overnight, we have gone from that to a couple of deployments per day. This increase means we need a preview environment per pull request, and this ties back to my question: how is CI keeping up?
Digging further
After the netlify incident i took a closer look at the other parts of my pipeline and and noticed we were also approaching sub $100 build minutes on GitHub, again this is shocking because for th most part we have kept CI pretty lean with tests, security scans and a preview environment on netlify where necessary.
To combat this, I’m spending the coming week auditing our deployments and looking closer at the tools we use and ensuring we don’t get a surprise bill. I suppose you could call this growing pains or just the pain of adopting DevOps practices.
In any case, you should too.
⚙️Tools I found useful
On a lighter note, this past week was in part filled with discovering a few small tools that are useful enough; I thought I’d share.
I’ve been trying out Herdr; as a long-time tmux user, I think this is a multiplexer that finally comes close. I have tried zellij in the past, but it never really stuck. Herdr also supporting tmux prefix and keybinds out of the box significantly reduced the learning curve.
I also discovered Kavita, which is a self-hosted digital library that I plan to run in my homelab to help organize my e-book collection. Definitely check it out if you have been planning on doing the same.
Tether was an interesting project that came up on my radar as it aims to support all of the Apple Continuity features on Linux.
🔥TOGETHER WITH EVERYTHINGDEVOPS
Everyone's asking if AI will replace DevOps engineers. That's the wrong question. The sharper one: what happens when a single engineer can manage ten times more infrastructure because agents are working alongside them?
In this video, Divine breaks down the 70% of your current skillset that still matters, the 30% that opens the real opportunities, and why AI-generated Terraform makes deep fundamentals more valuable, not less.
You cannot review infrastructure you don't understand.
He also lays out three projects you could build this weekend to prove the skills, not just claim them.
Subscribe: @EverythingDevOpsHQ for deeper dives on agents, Kubernetes, MLOps, and AI infrastructure.
🌍IN THE ECOSYSTEM
Really liked this blog: on the harness is the thing which echoes some of my thoughts from a previous issue on harnesses
Kubernetes 1.37 brings a much awaited feature which I explained over on LinkedIn
And for everyone who liked Go, I read this blog on why precision is the missing virtue in Go interfaces
⏱️UNTIL NEXT TIME
The pace at which we ship software these days truly requires a second look and a better review process; the tooling which I have relied on for years (paid or free) without having to bat an eye suddenly requires me to check if I'm exceeding their limits.
If you take anything away from this issue, it is to check on your pipeline; it deserves some love too.
Know an engineer wrestling with this? Share this link with them
Happy Monday!
Jubril Oyetunji
CTO, EverythingDevOps