Understand
Find the system underneath the symptoms.
Cincinnati
Toni Hale
I turn complicated systems into clear decisions, useful tools, and understandable stories.
Analytics, healthcare, project systems, and education are different lenses on the same problem: how do we understand complexity well enough to make better decisions?
- Understand
- Structure
- Analyze
- Communicate
- Improve
Capability map
Select a node
Complex Systems: the middle of the map. Select a capability to see the specific places it shows up.
Healthcare
Operational systems, specialty practices, and the information that moves through them.
Analytics
Turning raw information into measures someone can define, inspect, and use.
Project Management
Converting an unclear request into coordinated work.
Consulting
Diagnosing a problem and structuring a response people can act on.
Education
Helping someone else build a mental model they can use without me in the room.
AI
Learning emerging technical systems from the mechanism, not the slogan.
Information Systems
How data, definitions, and reports connect, and where they quietly diverge.
Communication
Compressing a complicated situation without sanding off the part that changes the decision.
01. What I do
Understand, structure, explain
Three moves, repeated across healthcare, analytics, delivery, and teaching. The links go to evidence.
Structure
Turn ambiguity into models, requirements, workflows, and measurable relationships.
Explain
Translate the result into something useful to executives, analysts, students, clients, or builders.
02. Selected work
Evidence, not a slogan
Independent venture
Independent Practice Resilience Assessment
A structured way for an independent practice to see where it is strong, exposed, or flying blind, before anyone proposes a system rebuild.
- ProblemOwners feel payer, administrative, and staffing pressure before they can point to a shared fact base.
- RoleFounder. Designed the assessment model, the public explanation, and the boundary of what the work will not claim.
- ApproachNine domains, one operating cycle: Assess, Build, Operate, Improve. Technology stays subordinate to the decision.
- ResultA published methodology and intake path. Measured client outcomes are not claimed.
- DemonstratesProblem framing, methodology design, and scope that stays honest.
Employment and consulting
Requirements and delivery in specialty care
Project, requirements, and analytics work across Eye Care Partners, Conclusn, and Parable, described at the level confidentiality and the public record allow.
- ProblemRequests arrived as symptoms. The work was to find the decision, the people, and the information the decision actually depended on.
- ResultCoordinated delivery and analysis shaped for the people who had to use it. Client-confidential metrics stay on the resume PDF, not as invented case-study numbers.
- DemonstratesAmbiguity reduction and stakeholder translation in real operating environments.
Teaching
Financial modeling and Power BI, taught
Adjunct teaching at the University of Cincinnati: graduate financial data visualization with Power BI, aimed at fluency rather than a tour of features.
- ProblemA model that only its author can explain is a fragile operational asset.
- ResultGraduate coursework in financial data visualization. Student performance metrics are not claimed.
- DemonstratesInstructional design and the same translation skill used with clients, pointed at learners.
Delivery context and venture method
Revenue and operations, made inspectable
The recurring analytics problem in specialty practices: reports exist, but the path from source event to a number leadership will stand behind does not.
- ProblemTwo reports about the same operation disagree, and nobody can say which definition the decision should use.
- ResultClearer definitions and reporting structure where the work was commissioned. No performance metrics are published.
- DemonstratesSource-to-decision thinking, KPI design, and restraint about what a chart proves.
03. Expertise
Claim, then evidence
Open a domain to see the work behind it. The longer pages go one level further.
Healthcare
Specialty operations, revenue cycle concepts, and the reporting practices use to see themselves.
Analytics and BANA
Business analytics as a discipline of definitions, models, and decisions, not a stack of chart types.
Consulting
Discovery, framing, and recommendations that a client can carry after the meeting.
Project management
Scope, coordination, and requirements in environments where the request starts out ambiguous.
Information systems and data architecture
How a fact moves from a source system to a sentence in a meeting, and where it changes meaning on the way.
Executive communication
Progressive compression: evidence, analysis, finding, decision, summary, each level honest to the one below it.
Education
University of Cincinnati teaching, and the broader practice of building a mental model someone else can use.
Artificial intelligence
Transformer inference, studied from the mechanism and taught at more than one depth.
04. A door into the mechanism
From Prompt to Token: How an LLM Generates a Response
A visual walkthrough from raw text to the next generated token, and why inference behaves differently during prefill and decoding.