21 Years Mastering What Matters
Patrick Dean Fox
Analytics Implementation Manager
Full-Stack BI Developer • Data Engineer • AI-Augmented Builder
“A jack of all trades is a master of none, but oftentimes better than a master of one.”
— The full quote that defines my approach
I believe in mastering whatever I invest my time in. Whether it's finance, operations, data engineering, or AI—I dive deep, learn fast, and deliver. My diverse background isn't a distraction; it's my superpower for solving complex, cross-functional problems.
Portal Pages Built
Datasources Integrated
Client Portals
Consulting Hours
The short version
Three numbers, each one auditable.
Every figure on this site is labelled by how it was derived, and every estimate shows its arithmetic. There are 75 of them in total — these are the three worth leading with.
~1,480
hours of manual migration absorbed
A platform-wide API deprecation across 40+ production portals, replaced by tooling I built. Labelled as the baseline addressed, not net hours saved — the post-automation cost was never measured.
See the evidence →
1,327
tests behind the agent tooling
Four open-source MCP servers that let an AI operate a production analytics platform, with write-safety gated by risk domain and every content write mirrored to git.
See the evidence →
24+
client portals built
Plus 200+ pages, charts and dashboards. Ten years of operations leadership before this, which is why the dashboards get opened.
See the evidence →
AI & Automation
AI tools have genuinely changed how I work. Not replacing what I do, but letting me do more of it, faster, across a wider range of challenges.
The AI Multiplier
13 years of domain knowledge combined with modern AI assistance. I know what to build and why—AI helps me build it faster.
Faster Delivery
Projects that used to take weeks can happen in days. Not because I'm cutting corners, but because AI handles the tedious parts.
Broader Capabilities
I can now take on front-end work, write Python scripts, and build things I would have delegated before. AI fills in the gaps.
Vibe Coding
Describe what you want, iterate quickly, ship something. It's a different way of working that feels more like collaboration than solo development.
Less Grunt Work
Boilerplate, documentation, refactoring—AI handles the stuff that used to eat up time. I focus on the actual problem.
Local LLM Experiments
Playing with local language models to understand what's possible. Data privacy matters for enterprise use cases, and running things locally gives you more control.
Chatbot Development
Building chatbots that actually connect to business systems using MCP (Model Context Protocol). Not just chat interfaces, but agents that can pull data and take actions.
Prompt Engineering
Learning what makes prompts work well. It's part art, part science. Good prompts can be the difference between useless output and genuinely helpful AI assistance.
AI in Workflows
Finding places where AI can actually help in day-to-day work. Not everything needs AI, but some tasks—like drafting, summarizing, coding boilerplate—really benefit.
This portfolio was built with AI assistance—a demonstration of the workflow
The rest is evidence.
12 case studies with their sources named, a career that ran operations before it ran analytics, and the things I build when nobody is grading them.