What I work on
I design and maintain cloud infrastructure across a large multi-account AWS environment: Terraform for provisioning, Spacelift for the review and apply workflow, ECS and Fargate for the services themselves. I build and maintain the delivery pipelines those services ship through, and I carry the observability and incident-response side of that as well.
The through line is that infrastructure should be describable. If the only record of how production got this way is somebody's memory, the system is already fragile, whatever the uptime graph says.
How I got here
I started as a data analyst, building predictive models and web scrapers to answer business questions. Then I spent a stretch as a marketing director, which is where I learned to index and analyze competitor sites with Elasticsearch, and where I found out I would rather build the tool than present the slide about it.
Since 2021 I have been doing infrastructure full time. The analyst habit never really left: I still reach for measurement before opinion, I am suspicious of dashboards nobody has validated, and I would rather find the number than argue about the vibe. That is most of what my engineering strategy comes down to.
I also freelance, and have since 2017, mostly building data collection and automation tooling for clients who need a question answered on a schedule rather than once.
What I write
Most of this site is writing rather than resume. There is the strategy, which is the set of trade-offs I work from; five playbooks covering how I actually use Terraform, containers, pipelines, observability, and Python; and a page on working with Claude, which is the honest version rather than the demo version.
Writing it down is not incidental to the job. A runbook nobody can follow and an argument nobody can check are the same failure.
What I read
A lot of what I believe about this work came out of books rather than out of incidents, and it is only fair to say which ones. I keep a list of the books that moved a specific position, each pointing at the page here where that idea actually shows up. Goldratt and Gawande earned their places as much as the Google SRE book did.
The book
Outside of infrastructure, I am writing a book called The Next 50 Years. It is about what happens when everything we built on growth runs out of people: the demographics, why the trend resists reversal, and what falls out of it. The second draft is online and readable in full.
The Next 50 Years — second draft, free to read
What happens when everything we built on growth runs out of people.
It is a different subject from the day job, but not a different habit. It is the same instinct that makes me want the state file to be accurate: work out what is actually true, write it down in a form somebody else can check, and be willing to be wrong in public.
Getting in touch
I am always happy to talk about infrastructure work, and I read everything that comes in. The fastest routes are LinkedIn and email.