Welcome to E-MyR

Evidence of Impact

• Designed reusable Epic documentation components, including SmartPhrases, SmartLists, and signature workflows, adopted across a 30+ clinician team and estimated to reduce charting burden by about 1 hour per clinician per week
• Standardized high-volume triage documentation structures to reduce variability, rework, and downstream interpretation burden
• Managed 80-100 patient encounters per shift while maintaining clinical documentation accuracy and safe escalation patterns
• Built governed local-first prototypes for document intake, review-safe LLM assistance, operational memory, and workflow observability

History - Systems I Have Worked Inside

TL;DR: Clinical triage, emergency care, rehab operations, EHR documentation, local-first software, governed AI, biological systems, ecommerce infrastructure.

My work is systems-first because my background has been systems-heavy. I have worked inside clinical environments where unclear handoffs create risk, remote workflows where documentation becomes infrastructure, and technical projects where a system is only useful if it behaves safely under constraint.

Clinical workflow and documentation systems

Most of my recent clinical work has been in remote triage, chronic care, and documentation-heavy healthcare operations.

• Mapped nurse-led triage and care-management workflows into clearer decision lanes
• Designed structured documentation components that reduced repeated entry and improved consistency
• Identified friction between nurses, providers, administrative teams, and downstream interpretation needs
• Worked directly inside EHR and remote-care systems where the practical question is not whether a tool exists, but whether it supports safe work

Emergency, rehab, and high-acuity environments

Earlier clinical work in emergency and post-acute settings shaped how I think about handoffs, failure modes, and cognitive load.

• Worked in environments where unclear lanes, incomplete information, and timing pressure directly affect safety
• Saw how small workflow defects compound during high-stress transitions of care
• Learned to treat handoff quality, escalation logic, and documentation clarity as operational safety issues, not administrative preferences

Technical and governed AI systems

In parallel with nursing, I have built software systems that explore how AI and automation can support judgment without becoming unaccountable infrastructure.

• Built local-first tools for document intake, knowledge organization, operational memory, and review workflows
• Designed governance patterns for human-reviewed AI output, audit trails, state visibility, and boundary enforcement
• Used Python, JavaScript, TypeScript, SQL, FastAPI, React, Vite, Pydantic, Pytest, Playwright, Git, GitHub, SQLite, and REST APIs across applied prototypes
• Focused on tools that expose uncertainty, provenance, and review status rather than hiding them behind a fluent interface

Experimental and creative systems

I also build outside traditional software categories. That work informs how I think about resilience, feedback, constraints, and interface design.

• Built and tested small biological systems including hydroponics, aquaponics, nutrient-flow experiments, and regenerative garden layouts
• Developed ecommerce and creative infrastructure for publishing, product workflows, automation, and visual identity
• Used photography, macro videography, and visual storytelling to study signal, attention, pattern, and perception

Organizations and contexts

A non-exhaustive list of organizations whose systems and workflows have shaped how I think:

Healthcare and clinical systems:

• Signallamp Health / TelliHealth
• Nobis Rehabilitation Partners
• Encompass Health
• Vitalounge
• Quantum Health
• Emergency and travel nursing contexts

Creative and systems-adjacent contexts:

• Independent software prototypes
• Visual media and ecommerce systems
• Local-first knowledge and workflow tools
• Biological and garden-system experiments

Diagnosis - Problems I Am Built to See

This section is where projects and case studies live. The common thread is not the category of the project. It is the pattern of the problem.

Meds – Interventions I Reach For

TL;DR: Map the real workflow. Clarify the lanes. Structure the documentation. Preserve human judgment. Add automation only after the system can be evaluated.

I do not treat AI as the intervention by default. Often the useful intervention is upstream: naming the workflow, reducing ambiguity, strengthening documentation, and deciding where automation is safe enough to test.

Workflow and lane clarity

• Map the actual workflow rather than the policy version
• Name decision points, handoff points, and ownership boundaries
• Separate nurse-resolved work, provider-escalated work, administrative work, and automation candidates
• Reduce cognitive ping-pong between teams by clarifying what happens next and who owns it

Documentation and signal design

• Refactor templates, SmartPhrases, and structured prompts around how clinicians actually reason
• Reduce duplicate entry and unnecessary variation
• Improve the signal quality of notes, triage summaries, handoffs, and downstream review artifacts
• Design documentation as operational infrastructure, not clerical residue

AI and automation boundaries

• Identify safe AI touchpoints for drafting, summarization, intake support, routing, and review assistance
• Require human review where clinical judgment, accountability, or patient safety are involved
• Make provenance, uncertainty, source state, and review status visible
• Define failure modes before expanding automation

Interface and system design

• Reorder fields, sections, and prompts around user reasoning rather than database convenience
• Surface next-best-action cues without pretending the system can replace judgment
• Reduce hidden friction from extra clicks, unclear states, ambiguous alerts, and scattered context
• Build tools that can be inspected, tested, revised, and governed

Labs – Signals I Watch

TL;DR: Friction, rework, handoff quality, cognitive load, provenance, review status, and how the system behaves under stress.

A system can look successful while quietly transferring burden to the people inside it. These are the signals I watch before trusting the story the dashboard tells.

Friction and rework

• Repeated clarification requests
• Duplicate documentation of the same clinical story
• Message reroutes, wrong-team handoffs, and unresolved ownership
• Shadow spreadsheets, informal workarounds, and local memory systems required to keep work moving

Handoff quality

• Whether the next person can act without reconstructing the whole situation
• Whether escalation criteria are explicit enough to be used consistently
• Whether context survives transitions between nurse, provider, admin, and downstream review teams
• Whether handoffs reduce uncertainty or merely move it elsewhere

Cognitive load

• How much attention goes to the patient or task versus remembering how to operate the system
• Whether the workflow feels like a coherent story or a puzzle
• Whether alerts, inboxes, forms, and dashboards create signal or noise
• Whether the system relies on heroic memory instead of good design

AI governance and reliability

• Whether AI output is sourced, reviewed, and bounded
• Whether the system distinguishes confidence from evidence
• Whether uncertain outputs remain visibly uncertain
• Whether automation creates accountability gaps
• Whether failure states are explicit, logged, and recoverable

Operational resilience

• How the system behaves during surges, interruptions, staffing pressure, and policy changes
• Whether small defects are surfaced early or allowed to accumulate invisibly
• Whether changes can be tested without destabilizing the workflow
• Whether the system can explain what happened after the fact

Notes - Progress Notes and Field Observations

TL;DR: Short essays and working notes on clinical systems, AI governance, documentation, workflow geometry, cognitive load, and the strange middle ground between disciplines.

The notes section is not a traditional blog. It is closer to a public notebook: observations, working language, project reflections, and attempts to name patterns before the categories fully settle.

Documents – Verification

TL;DR: Resume, credentials, source links, and professional verification.

I keep public claims conservative. Where possible, I prefer concrete artifacts, visible systems, and bounded evidence over inflated positioning.