I had a question. So I built a world where I could find out.

I build simulated systems to investigate how data, infrastructure, and organizations behave when reality stops cooperating with the design.

My background is in healthcare operations, where I spent years working across the workflows, systems, data, and people behind patient access and clinical operations.

Now I use that experience as a starting point for building, breaking, investigating, and rebuilding technical systems.

What happened?

Investigating data across systems, reconstructing events from incomplete evidence, testing assumptions, and determining what can actually be proven.

Current investigation: The Data That Refuses to Die — in development.

A third-party AI system has supposedly been decommissioned and its customer data removed.

The assignment is simple:

Prove it.

What did we inherit?

A longitudinal experiment in inheriting an organization whose original technical team is gone but whose systems are still running.

The work: discover what exists, reconstruct dependencies and ownership, determine what can be trusted, and decide what can be safely changed.

How do we keep it operating?

Experiments in infrastructure reliability, capacity, power, cooling, workload prioritization, failure recovery, and the data required to understand what a facility is actually doing.

How do we protect and modernize systems society cannot afford to lose?

Long-horizon investigations into legacy infrastructure, IT/OT dependencies, resilience, modernization, data architecture, governance, and emerging technology risk.

The lab follows a simple cycle:

Build → Observe → Break → Investigate → Remediate → Retest → Document

A system working is not enough.

I want to know:

• What happens when an assumption is wrong?
• What evidence proves the system is behaving as expected?
• What survives when a component fails?
• Can another person understand what happened?
• Can they safely inherit what remains?

Documentation is part of the system.

The lab didn't start here.

Earlier projects remain in this repository history as evidence of the learning process — including SQL investigations, healthcare data analysis, workflow automation, debugging, incorrect assumptions, corrections, and progressively more complex system work.

• Jira Ticket Triage Bot
• AI Sprint Report Generator
• Confluence Weekly Digest

These projects explore APIs, orchestration, structured data, AI-assisted processing, and automated reporting.

SQL · Python · REST APIs · JSON · Git/GitHub · Linux/Bash · Databricks · Jira · Confluence · Make

Tools are added when the problem requires them.

AI may be used as a simulation engine, synthetic artifact generator, or technical tutor.

When used in an investigation, it does not make the investigative decisions for me.

The hypotheses, queries, evidence assessment, remediation decisions, and conclusions are mine.

Service. Stewardship. Trustworthy AI requires trustworthy data.

📍 @cocoagadget