jr-articles
Writing and editorial source.
About this site
I lead technology, and I still build because direct contact with systems improves my judgment about architecture, security, operations, teams, and investment.
This site is one of those systems. Trellis is a working laboratory for decisions about trust boundaries, cost, automation, observability, and sustained operation. It includes the public site, security monitoring, analytics, and the infrastructure used to run them.
How the pieces fit together
Trellis
Public site and security event collection.
Celbase
An anonymized view of activity observed by the system.
rehagen-ops
Monitoring, analytics, alerting, and operations.
How I approach it
Separate collection from analysis
The public system collects events. Private systems handle monitoring and analysis. They have different users, risks, and deployment cycles, so they remain separate.
Automate changes, retain judgment
Infrastructure and deployments are automated and reviewed. Changes with meaningful operational consequences still require an explicit decision.
Operate within constraints
Cost, security, recovery, and observability are part of the design. The system is intentionally small. I document decisions and automate routine work so it still makes sense months later.
What this is—and isn't
This is a personal system operated by one person. It uses production practices where they improve reliability, security, or understanding, but it is not a model for running an enterprise environment.
The scale is deliberately small. Some choices optimize for cost or learning rather than availability. The system changes as I learn from operating it, and the writing records those changes—including the decisions that did not hold up.
I use AI extensively to research, write code, review changes, and test assumptions. I remain responsible for the architecture, decisions, validation, and operation of the system.
See the system
Explore presents an anonymized view of the activity the system observes →