I’m a software engineer and systems architect with nearly two decades of experience building and modernizing complex systems across government and private-sector organizations.
My work spans generations of technology — from mainframe systems to cloud-native platforms — but the focus has always been the same: designing systems that are reliable, scalable, and built to last. I care about clear boundaries, strong data models, and architectures that hold up under real-world pressure.
I work across languages and platforms as needed — including C#, Java, and JavaScript — but I don’t define myself by a stack. The problem space drives the tools, not the other way around.
I treat AI as a force multiplier, not a replacement for engineers. It’s a capable junior developer: fast and tireless, but still in need of direction. I use it to accelerate execution while I focus on architecture, intent, and correctness. Writing code isn’t the job — engineering reliable systems is.
Migrate legacy workloads to Azure or AWS with Infrastructure-as-Code, automation, and observability built in.
Design clear service boundaries, auth flows, and gateway policies for distributed systems that are easy to reason about and operate.
Build reliable ETL/ELT pipelines, APIs, and event-driven integrations using ADF, SSIS, PostgreSQL, and SQL Server.
Help teams adopt better patterns, tame complexity, and ship systems that are maintainable long after the first release.
I use AI as a capable junior developer — fast, literal, and tireless — while I stay accountable for architecture, security, and correctness. My focus is on using AI to accelerate delivery, not to replace engineering judgment.
If you’re curious about how I think about this in practice, you can read my Principles of AI-Augmented Engineering on my GitHub profile.
No skills match that filter.
A detailed version of my career history — including key projects, leadership, and modernization efforts — is available in my full career archive.
2021 – Present
JBS.dev
Consulting across architecture, backend, DevOps, and data engineering — with a focus on cloud-first, secure, and reliable systems.
2016 – 2021
U.S. Department of the Treasury, Bureau of the Fiscal Service
Led modernization of public-facing federal financial systems and identity platforms, bridging mainframe, Java, and Azure.
2007 – 2016
U.S. Department of the Treasury, Bureau of the Fiscal Service
Built and maintained core Treasury systems across mainframe and Java stacks.
C# / .NET Azure Functions • Angular • Azure SQL • ARM • FedRAMP High
Led the design and delivery of the Savings Bond Valuation & Verification (SBVV) tool for financial institutions nationwide. Built on Azure using serverless compute, Angular, and Azure SQL, SBVV enables tellers to instantly value and validate savings bonds and detect invalid or fraudulent instruments.
C# / .NET Azure Functions • Azure SQL • Azure API Management • Angular
Served as lead architect and developer for the Treasury Check Verification System (TCVS), created to combat fraud around U.S. Treasury checks, including COVID-era relief payments. Implemented a real-time verification API and web experience that helps financial institutions validate checks before acceptance, preventing large-scale fraud.
Azure App Service • Azure Functions • Azure SQL • ARM • Governance • FedRAMP High
Pioneered Azure adoption within the Bureau of the Fiscal Service, designing, operating, and governing its first production Azure environment. Migrated billions of savings bond records from DB2 to Azure SQL, implemented Infrastructure-as-Code with ARM, and managed platform security, access, and policy under FedRAMP High requirements.
Java 17 / Spring Boot • AWS ECS • PostgreSQL • CI/CD • IoT & reporting
For an energy-sector client, delivered major features for a sustainability and emissions tracking platform, including emissions goal modeling, rate calculations, and real-time meter data ingestion. Led the migration of the platform to AWS ECS with full containerization and one-click CI/CD, and designed PostgreSQL-backed reporting APIs with significant performance tuning.
Azure Data Factory • Azure Functions • SQL Server • Bicep • API Management • Key Vault
Architected and delivered secure Azure-based data platforms for financial workflows, replacing legacy ETL and batch processes with observable, automated, cloud-native services. Designed partner data-exchange patterns, rewards-processing pipelines, source-data ingestion, reconciliation, replay/backfill utilities, and production-ready infrastructure using ADF, Azure Functions, SQL Server, Bicep, Blob Storage, API Management, Application Gateway, Key Vault, and private networking.
Chunking Collector is a small, focused Java library that makes it easy to turn streams and iterables into fixed-size chunks, sliding windows, and weighted batches — without pulling in a framework or runtime dependency.
IntStream, LongStream, and DoubleStream.
Compose Reorderable Grid is a production-grade, long-press drag-and-drop grid for Jetpack Compose. It provides smooth, predictable reordering with auto-scroll support and stable gesture handling, extracted and refined from a real shipping Android app.
Score Card is a research project and working prototype that explores a new way to manage complex, cross-service transactions without relying on heavyweight orchestration, 2PC, or tightly coupled saga coordinators. Instead of forcing microservices to participate in a global transaction, Score Card uses observability, correlation, and dependency-aware authorization to reconstruct and reason about multi-step workflows as they unfold.
At its core, Score Card models a transaction as a graph of actions across independent services. Each service performs its work autonomously and reports status updates back to the Score Card engine, which assembles a real-time view of the end-to-end flow. This enables failure detection, auditing, recovery, and compensation without sacrificing service autonomy or scalability.
ActionWatch is a platform for understanding what is really happening inside GitHub Actions. It turns webhook-driven workflow activity into live operational visibility, historical analytics, and dashboards that help teams debug failures, track reliability, and understand CI/CD behavior over time.
ActionWatch is also a showcase for AI-augmented engineering in practice: 100% AI-coded, with the architecture, intent, and final engineering judgment driven by me. I wrote about that workflow here.
I operate a personal Linux server hosting a suite of containerized services, including Nextcloud, Pi-hole, WireGuard VPN, and monitoring tools. This homelab serves as a sandbox for experimenting with container orchestration, networking, automation, and secure self-hosting — and keeps my personal infrastructure fully under my control.
Keller Graduate School of Management — Focused on information systems strategy, management, and the alignment of technology with business goals.
DeVry University — Emphasis on software development, object-oriented programming, and web technologies.
Washington State Community College — Foundation in programming, databases, and business computing.