Safe migrations
I’ve moved production workloads across virtualization platforms while protecting shared storage, source data, application dependencies, and rollback options.
About me
I’m Joseph Miller. I like the problems without an obvious owner.
I’m at my best when a technical problem crosses boundaries: identity touches networking, storage changes affect virtualization, a cloud symptom starts on a Windows host, or a migration is really a recovery problem in disguise.
I live in Oregon, and I spend a lot of my time building, troubleshooting, documenting, and learning how complex systems behave under real constraints.
How I got here
I did not follow one narrow product track. I kept following the problems. That took me through Windows Server, Active Directory, Microsoft 365, virtualization, SAN storage, networks, firewalls, backups, security monitoring, Linux, containers, automation, GPUs, local inference, and agent systems.
That breadth became one of my strengths. I can move between the architectural view and the low-level evidence: the migration sequence and the disk layout, the identity design and the token failure, the public service and the DNS, firewall, TLS, proxy, container, and application path behind it.
I care about safe execution because the most interesting systems are usually the ones people already depend on. I want a clear starting state, a recovery option, observable checkpoints, and proof that the complete function recovered—not just a green status beside the component I touched.
What Socket23 means
It’s my personal lab and public notebook.
Socket23 is the handle I use for my technical projects, lab work, notes, and long-term body of work. It gives me one place to connect infrastructure, cybersecurity, recovery, automation, and private AI.
The work here represents me: what I have built, what I have successfully worked through, what I am still learning, and how I think about risk.
What belongs here
- Projects I can explain honestly
- Sanitized lessons from production work
- Lab systems and experiments
- Technical notes and decision records
- New ideas with clear boundaries
Things I’m proud of
Success, to me, is more than getting past the error
Making old systems supportable
I’ve mapped and stabilized inherited Windows environments where the documentation did not tell the whole story.
Recovering the complete function
I look beyond backup-job success and validate boot, identity, data, applications, dependencies, and the actual user outcome.
Building my own platform
I built the public delivery system for this site and a private AI lab so I can test architecture, security, automation, and failure behavior directly.
How I work
Curious, careful, and evidence-driven
I verify the environment
I do not assume the system matches the diagram, ticket, memory, or original design.
I make risk visible
I identify the authoritative data, hidden dependencies, irreversible steps, and recovery options before changing the system.
I write down what matters
I document the observed state, the decision, the result, what remains uncertain, and the next safe action.
Education and credentials
I keep learning close to the work
My background includes cybersecurity education and verified credentials across monitoring, incident handling, network analysis, Python automation, cloud security automation, and machine learning engineering.
Credly is the current source for credential status rather than a static badge count on this site.