Experience
A few recent highlights from 20+ years of engineering work.
Software Engineer III
Overview
Technical lead for Georgia’s Customer Portal (gateway.ga.gov) and Worker Portal, the primary interfaces through which millions of citizens apply for and manage health insurance, SNAP, Medicaid, childcare assistance, and other critical services—and through which thousands of state workers process those applications. Joined Deloitte in 2022 as Project Delivery Specialist, promoted to Project Delivery Specialist II in 2025, and promoted again to Software Engineer III in 2026—a track reflecting steadily increasing ownership over architecture, security leadership, and cross-team technical direction.
My role blends architecture (30% of my time) with hands-on coding, security leadership, accessibility advocacy, production support, and cross-team coordination (70%). I make system design decisions affecting multiple teams, write code to accelerate delivery, establish patterns others can follow, coordinate vulnerability remediation, drive accessibility compliance, resolve complex production incidents, and serve as a technical bridge between developers, business analysts, testers, infrastructure teams, and state clients.
Key Accomplishments
Led design and delivery of Worker Portal View 360, a read-only, integrated 360-degree case and application experience built inside the existing Worker Portal without disrupting legacy operations. Served as Tech Lead and Architect across the full stack, aligning session/launch-flow patterns with Worker Portal’s existing security model. Piloted June 28, 2026 and scaled to ~800 daily users at ~80,000 requests/day with ~33ms average response time.
Led modernization of gateway.ga.gov using Georgia’s Orchard design system, resolving 200+ Snyk-identified critical JSP vulnerabilities, partnering with state ADA teams to validate WCAG 2.2 compliance, and delivering a fully responsive, mobile-first experience. This project standardized reusable components across Georgia state sites and drastically reduced attack surface.
Coordinated remediation of 1000+ critical and high-severity vulnerabilities reported by CISA, Snyk, SONAR, and HackerOne. Partnered with the security team to triage findings, coordinate fixes across development squads, validate remediations, and track progress to closure. Significantly elevated the security posture of both portals.
Designed and implemented duplicate detection framework using Oracle-based similarity scoring (SSN, DOB, gender, name matching via exact/phonetic/Levenshtein algorithms). Automated primary/duplicate designation, added Linked Application panel for workers, and reduced manual consolidation workload while improving data integrity.
Led Intelligent Task Routing (ITR) initiative across multiple squads, replacing manual/batch task assignment with real-time, priority-based routing. Redesigned architecture and data model for scalability, introduced workgroups for load management, implemented complexity-based assignment logic considering worker experience, added Active/Pending Dashboards, and automated reallocation for inactive users.
Championed accessibility across engineering org, serving as ADA leader. Conducted design reviews with accessibility lenses, partnered with business analysts and developers to understand WCAG 2.2 requirements, validated implementations with state ADA teams, and established component patterns that make inclusive design the default rather than an afterthought.
Established GitLab standards by co-leading initial setup and documentation, standardizing branching strategies, CI/CD pipelines, and release processes. Improved development team throughput and reduced friction during deployments.
Implemented emergency reCAPTCHA Enterprise v3 integration overnight to mitigate automated bot attacks targeting the Customer Portal. Coordinated with infrastructure and security teams to deploy under time pressure without causing downtime.
Business Impact
Reduced citizen support volume through transparent Application Status Tracker. Decreased administrative workload via automated duplicate detection. Accelerated case handling through real-time task prioritization. Improved audit readiness and compliance posture by rapidly addressing security vulnerabilities. Enabled accessible, mobile-first self-service meeting validated WCAG 2.2 standards. Delivered more predictable service levels and faster time-to-resolution for citizens applying for benefits.
Interesting Challenges
Navigating a decade of technical debt. The Customer and Worker Portals were built over 10+ years by dozens of developers with varying standards. Improving them requires understanding why decisions were made, identifying debt that matters vs. debt that doesn’t, and making surgical changes that deliver value without requiring ground-up rewrites.
Balancing security with deadlines. Security vulnerabilities arrive on their own schedule. Coordinating remediation across teams—each with their own sprint commitments—requires constant prioritization, clear communication of risk, and finding creative ways to batch fixes without compromising security.
Making accessibility real, not performative. It’s easy to run automated scans and fix low-hanging fruit. Harder to shift an organization’s mindset from “compliance checkbox” to “build inclusive experiences by default.” This requires patient education, design review engagement, and demonstrating—through working code—that accessible design isn’t extra work when it’s built in from the beginning.
Oracle at scale. Optimizing queries for millions of records in Oracle requires deep understanding of execution plans, index strategies, and when to push logic into the database vs. application layer. Building the duplicate detection similarity scoring function directly in Oracle dramatically improved performance compared to application-level processing.
Lessons Learned
Context switching is expensive. Balancing architecture, coding, production support, security coordination, and accessibility leadership requires protecting deep work time. I’ve learned to batch similar activities and communicate my availability clearly.
Documentation prevents repeated questions. When teams ask the same question multiple times, the problem isn’t people—it’s missing documentation. Investing time in clear, searchable docs reduces future interruptions and scales knowledge.
Production incidents reveal system gaps. Every incident is a signal. Not just “what broke,” but “what process failed that let this reach production?” and “what monitoring could have caught this earlier?” Blameless post-mortems turn failures into learning.
Security findings are not optional work. Treating vulnerabilities as backlog items to address “later” is a recipe for disaster. Later never comes. Address critical/high findings immediately, coordinate remediation, track to closure.
Accessibility cannot be retrofitted. Every time I’ve tried to “add accessibility later,” it’s been more expensive and produced worse results than designing inclusively from the start. It’s not extra work if you build it in from the beginning.
Client Impact
The Georgia Integrated Eligibility System directly impacts millions of citizens applying for health insurance, SNAP, Medicaid, childcare assistance, and other critical services. My work ensures they can apply using mobile devices, navigate the system with assistive technologies, trust that their data is secure, and receive timely updates on application status—ultimately improving access to benefits that materially affect people’s lives.
Senior Java Developer Lead
Overview
Java Developer and Technical Lead for the Georgia Integrated Eligibility System (GA IES), working as a contractor for Deloitte. Started as a developer, quickly promoted to technical lead responsible for Maintenance & Operations (M&O) and Change Request (CR) teams.
Primary focus: analyzing and resolving production incidents, coordinating fixes across functional and technical teams, setting priorities, and ensuring optimal solutions. Later became ADA leader, working with clients, business analysts, developers, and testers to improve system accessibility following WCAG and ADA guidelines.
Key Accomplishments
Established accessibility practice. Built the initial plan for improving system accessibility: analysis of current state, design change recommendations, creation of internal standards, and coordination across functional, development, and testing teams to deliver better user experiences for citizens with disabilities.
Production incident leadership. Analyzed complex multi-layer incidents (frontend, backend, database, integration), coordinated rapid diagnosis and fixes, and communicated status to stakeholders. Built reputation for calm, systematic problem-solving under pressure.
Adapted during COVID-19. When the pandemic created urgent new requirements, shifted 40% of time to hands-on coding to help teams meet accelerated delivery timelines for features like application deadline extensions, enhanced notifications, and worker capacity management.
Lessons Learned
Incident response is about communication as much as technical skill. Stakeholders need to know: what broke, what’s the impact, what’s the estimated fix time, what’s the workaround. Clear, regular updates reduce panic even when the fix takes time.
Accessibility expertise is rare and valuable. Most engineers haven’t used screen readers or navigated systems via keyboard only. Becoming the “accessibility person” created opportunities to lead, teach, and influence system design.
Java Technical Leader
Overview
Project/Technical Leader and Software Architect assigned full-time to SAT (Servicio de Administración Tributaria—Mexico’s equivalent of the IRS), participating in high-visibility projects and supporting day-to-day operations of the technology team as part of the “SDMA-3” government tender.
Key Accomplishments
Led development of Latin America’s first electronic notification system. Delivered—along with my team—a robust, maintainable web application with zero defects leaked into production. This system transformed how SAT communicated with taxpayers, replacing physical mail with secure electronic notifications, dramatically reducing delivery times and costs.
Rescued two failing projects. Took control of both, reorganized teams, guided developers through technical challenges, attended regular client meetings to re-establish trust, redefined deliverables and scope to align with realistic timelines, and brought both projects to successful completion.
Implemented client architecture guidelines. Translated SAT’s architecture standards into practical implementations, guided developers through patterns and best practices, composed technical documents following client methodology (MDAP), and inspected releases to UAT and production environments.
Delivered a portfolio of SAT modernization projects, each with distinct architecture and technical leadership responsibilities: Buzón Tributario (Electronic Mailbox—MVC refactor, 3-layer separation, database indexing for query performance); NyV — Notificación y Verificación (massive electronic-document signing via a multithreaded Java applet using Bouncy Castle/MD5/SHA-1, plus AXIS web services for cross-system notification lookup, leading 8 developers); Fiscalización Electrónica AGAFF (rescued a past-deadline project, managed two 10-person teams, redesigned the database and tuned queries via execution plans and SQL Tuning Advisor); and the Semáforo Aduanal Inteligente (rules-based passenger risk-classification system for customs, including the underlying database model). Also led the TCS–Softtek transition of the SDMA-4 contract, introducing a NoSQL-backed architecture for AGAFF’s data-loading module.
Lessons Learned
Project rescues require reorganization before coding. Failing projects rarely fail due to technical skill gaps. They fail due to unclear priorities, misaligned expectations, poor communication, or missing process. Fixing those comes first.
Client trust is earned through consistency. Regular status updates, honest assessments of risk, and delivering on commitments—even small ones—compound over time. Trust, once lost, takes months of consistent behavior to rebuild.
Zero-defect releases are possible. The electronic notification system’s defect-free production launch resulted from rigorous testing, clear specifications, strong QA partnership, and empowering the team to say “this isn’t ready” without repercussion.
For my full work history, speaking engagements, and recommendations, see LinkedIn.