Manual operational work
Repetitive copying, re-keying, and status chasing that consume time without creating proportional value.
AI AUTOMATION SPECIALIST · CHICAGO
I make AI work for business operations.
I bring clinical-grade process discipline to workflow automation—translating operational reality into reliable code, explicit controls, and systems people can trust.
8+years in regulated clinical operations
18nodes in a documented qualification workflow
33/33database tests passed for ExoCore OS
CLINICAL OPERATIONS
+SYSTEMS THINKING
+AI WORKFLOW ENGINEERING
PROBLEMS I SOLVE
Operational friction that software should remove.
I look beyond individual tasks to find the broken handoffs, hidden rules, and missing controls affecting the wider system.
Disconnected systems
Business data trapped across tools that do not communicate reliably or preserve one clear source of truth.
Fragile workflows
Automations that depend on ideal inputs and fail without clear recovery paths when reality changes.
Invisible execution
Processes with limited monitoring, unclear ownership, and no usable record of what happened or why.
Undocumented logic
Critical rules held in people’s heads instead of being expressed in maintainable processes and software.
THE BRIDGE
Operations taught me where systems break.
Clinical research made process discipline non-negotiable. Work must be traceable. Exceptions must be visible. Human accountability cannot disappear behind software.
I now apply that operating mindset to AI automation—starting with the process, encoding business rules in software, and validating the system before calling it reliable.
WHY MY BACKGROUND MATTERSEight years in regulated clinical research taught me that systems must be accurate, traceable, documented, and accountable. I bring that same discipline to AI automation.
Process intelligence
Current-state mapping, waste diagnosis, requirements, risk, and future-state design.
Workflow engineering
n8n, APIs, JSON, Supabase, structured model outputs, routing, and failure recovery.
Governed AI systems
Human approval boundaries, audit trails, data controls, tests, and observable execution.
Clinical operations
Study startup through closeout, CTMS/eTMF, monitoring, data quality, and inspection readiness.
SELECTED WORK
Systems, not demos.
Each case study shows the business logic, implementation evidence, and current proof boundary.
View case study ↗ExoCore OS — Client Zero
- Operational problem
- Operational decisions, approvals, evidence, and lessons learned can become scattered across disconnected tools.
- System engineered
- A working operating console connecting process discovery, waste findings, initiative approval, KPI measurement, knowledge capture, and decision history.
- Controls implemented
- Founder-gated approvals, database policies, application checks, and an auditable activity trail.
Evidence produced
- 33/33 database tests
- 4/4 application tests
- Approval bypass rejected
View case study ↗Lead Qualification System
- Operational problem
- Manual lead review slows response, duplicates work, and makes qualification decisions difficult to apply consistently.
- System engineered
- An 18-node qualification architecture covering validation, email verification, AI scoring, enrichment, persistence, and tiered routing.
- Controls implemented
- Explicit failure alerts, duplicate routing, structured outputs, and retained processing records.
Evidence produced
- 18-node export
- 57 test records merged
- 42 duplicates routed
View case study ↗Understand Before You Automate
- Operational problem
- Automation often begins before the underlying process, waste, value case, and accountability boundaries are understood.
- System engineered
- A repeatable methodology for observing work, quantifying waste, redesigning the process, validating value, and then implementing automation.
- Controls implemented
- Human accountability, value validation, measurable outcomes, and continuous monitoring are built into the method.
Evidence produced
- 8-stage methodology
- 8 waste categories
- Human accountability
HOW I WORK
Understand before you automate.
AI is not the starting point. The work begins by understanding how value moves through the business and where waste, risk, and unclear decisions interrupt it.
- 01
Observe
See how work actually happens.
- 02
Measure
Quantify the waste and baseline.
- 03
Redesign
Simplify before adding software.
- 04
Validate
Prove value and protect judgment.
- 05
Build
Encode the approved logic reliably.
- 06
Improve
Monitor evidence and iterate.
EXPERIENCE
From trial operations to AI systems.
MBA, Healthcare · Careerist AI Automation Specialist Program
AI Automation Specialist · Maximax Automation Agency
Mapped manual workflows and operational bottlenecks, built n8n and Make.com automations using the OpenAI API, and documented system performance, error handling, and client handoffs.
Senior Clinical Research Associate · Karyopharm Therapeutics Inc.
Manage clinical-trial operations from planning through close-out, coordinate timelines and compliance checkpoints, maintain CTMS and eTMF records, resolve deviations and data discrepancies, and produce stakeholder updates.
Clinical Research Associate I & II · ProQR Therapeutics
Served as the primary contact for external sites, reviewed data for accuracy and completeness, resolved and escalated discrepancies, and delivered process training and guidance materials.
Clinical Research Coordinator · MD Anderson Cancer Center
Reviewed candidate records against defined criteria and maintained accurate, structured intake and electronic documentation supporting downstream reporting.
WORKING STACK
OpenAICodexn8nSupabaseNext.jsPostgreSQLGitHubAirtableObsidian
LET'S BUILD SOMETHING USEFUL
Looking for an operator who can think in processes and build in systems?
I'm open to AI automation, clinical technology, operational excellence, and systems implementation opportunities.