With all the attention frontend gets in tech, there’s still no better signal of an engineer’s crackedness than a good old home lab.
And honestly, if the CEO of your DevOps platform doesn’t run one, you’re cooked.
Am I wrong?
(P.S. this cable job is about on par with a data center)
Calling St. Louis makers and innovators. Building from zero is thrilling and lonely. We are quietly creating a space for people like us. If this resonates, reach out or comment here and I’ll reach out.
It is simple. People express their intent. The system handles the how.
Whoever builds the most human, intuitive way to do that will win.
And it will not be AI SREs.
I am tired of all the little “operation” buckets.
Tired of pretending we are saving the world when half the day is meetings about who broke prod or bouncing around Slack threads.
So what is the real end goal?
We are entering an age where billions of tiny applications will be built and run by people who do not care about titles like SRE or DevOps or SOC...or whatever
They just want their software to run and do the job.
Honestly, I want that too.
“Developers are frustrated with production. Hire someone for that.”
That worked when software was slower and centralized.
But that world is ending.
When I first started OpsCompanion, I used that language too. It felt familiar. But the truth is I was not thinking big enough.
These are features, not companies.
They come from an old era of software where every new problem created a new role.
AI DevOps. AI SRE. AI SOC. AI whatever operations term you want.
I understand why these exist. I understand the need, but I cannot stand the implementation.
I am super bummed I didn't hear about this till it was too late.
How did you hear about it? I need to change what I am indexing on
Cloudflare's engineers are some of the best in the world at piecing that context back together. The importance of software in 2025 is demanding something new.
If AI can understand human intent, and tell the real story of what changed
I think it can reshape reliability forever.
Software is a black box that will run forever until we poke and prod at it. And we poke at it constantly.
Code changes are the only thing we track cleanly with Git. Everything else that shapes a system such as configs, flags, infra tweaks, scripts, and Slack threads rarely gets written down.
Because every engineer who has lived through a real outage knows the truth:
You don’t debug the system.
You debug the last human decision that touched it.
So how do you debug a single piece of software, much less the systems that power the Internet?
You start with the change, then the context behind it.
In light of the Cloudflare outage yesterday, it is a good time to talk about how an engineer even begins to solve a problem like that.
When I was an SRE at Cloudflare I answered the pager hundreds of times, but nothing will ever match Code Orange or Code Red in 2023.
The heated steering wheel is such a hard thing to appreciate till you have it, then its incredible.
First meeting: @mohistorymuseum
Next stop: @STLArtMuseum
@OpsCompanion St. Louis history in progress.
A often forgotten concept in the software world is giving back the systems we maintain to our colleagues.
We should be striving for all engineers to be better stewards of production.
This is the future we must achieve.
Because in reality a lot of SRE and Devops teams have become designated ITOps teams that are tasked with dealing with ever growing bitrot in software.
We have also seen a massive rise in AI SRE tooling which again is another false hope.
For those who may not know, Google invented this role around 2003 to deal with their massively growing infra.
The unfortunate truth though, is that for most teams the lessons in these books just come to represent a false reality. Again, because most teams are not Big Tech.
Look, I loved being an SRE. The shared bond and level of talent between these engineers is unmatched in the tech industry.
But we need to start to ask ourselves if this role is really the answer to dealing with all of problems in production. Because most teams are not Google.
And if you think success is just raising capital, you’re wrong, it’s building a damn good business.
We went full-time at the end of August. Already at $60K ARR, with inbound from some of the biggest names out there.
Building a startup in the Midwest should be a requirement for most @ycombinator startups.
99% of the noise gets filtered out here.
If you’re not talking to users or selling, you die. Nobody cares about your tech if it doesn’t move the needle.
I've started working on working on improving the lives of SREs around the world, but I need your help. Please take a few minutes this week to capture your pains, dreams, and goals of being an engineer working on the world's hardest problem.
To the future: forms.gle/7fp1wTRoWQ75...
FRIDAY ALERT! Before you mentally check out, take 2 mins to fill this form. We're transforming DevOps & SRE practices - be part of the future of infrastructure automation! #CloudNative
docs.google.com/forms/d/e/1F...
#DevOps #SRE
After leading some of the world's largest systems, I've identified critical gaps in our industry. I'm building solutions, but I need your insights to make a broader impact. If you have a few minutes, please share your experiences: forms.gle/G8aRkXzpj6xH...
#devops #sre #SystemsEngineering
This team is cooking up something great
opscompanion.ai 👀
From Champions League to STL ✨
Watch the full profile on YouTube and get to know our German forward Cedric Teuchert.
📺 youtu.be/ehRddcnSq2I
Arnold Britt has been hired by the city’s troubled Building Division, which is under FBI investigation.
Does anyone else have those days as a software engineer where you just walk away from your computer both amazed at the technology and the sheer amount you need to learn?
Contact your Alder A.S.A.P!
#VoteNoOnBB161
@stateofstlouis.bsky.social breaks down the entitlement tax break for Anheuser-Busch, including prior incentives from the tax break buffet.
My email follows in next post.
#BOAbb161
/1
