I deploy and run Aceso Sphere, our flagship patient-care platform, across customer sites, each a five-to-six server footprint spanning app, admin, integration, database, video, and edge servers. Aceso Sphere runs on both the servers and the in-room TVs through Upserver, its .NET server application, which I deploy on IIS over an MSSQL backend I provision per site. Releases ship through an Octopus Deploy and GitLab pipeline I maintain, and a few older sites still run the video component on RHEL alongside a legacy MySQL instance.

I run the systems the platform depends on: a Mirth Connect interface engine feeding HL7 ADT and Epic integrations, an on-prem Redis instance for Entity Framework caching and pub/sub (a single instance, by design), production Pexip video conferencing, and a legacy Razuna deployment kept patched and current, including a production CVE I cleared by modernizing its CFML runtime onto current Java.

The underlying infrastructure is mine as well. The servers we host ourselves run on XCP-ng, which carries the work servers, lab, demo, and build environments, and the Pexip nodes, with AWS used in the office. I built the reporting automation that moves monthly site data into Google Drive, a PowerShell and Python uploader with OAuth and deduplication that runs unattended over WinSCP.

I monitor the deployed endpoints, primarily the in-room TVs that can number in the hundreds per site, work production-down issues directly with hospital teams (often through the MSSQL data to find the cause), and keep software and dependencies current across everything we deploy. It is an on-call, HIPAA-compliant environment held to a high bar for uptime and data handling.

As a senior member of the Global Cloud Services team, I was the liaison for hosted-solution customers across multiple verticals. I built long-term relationships with them and tailored changes and enhancements to each one's needs. I built and ran the cloud server infrastructure those clients relied on, including the HL7 and FHIR integration servers for healthcare customers, and managed the solutions running on OnBase, Hyland's content services platform, across both the infrastructure and the application layer.

I implemented OnBase solution modules to documented requirements, from new installs and upgrades to fit assessments, and carried customers through testing and the go-live migration into production. I owned a high volume of customer requests end to end, and implemented and verified load balancing, firewall, and proxy changes across test and production systems.

Day to day I monitored those environments for uptime and performance, triaged alerts across the hosted fleet in Splunk, reported through Power BI, and collaborated across management levels and with peers to onboard new customers and resolve the harder client-facing problems.

A planning tool for Redis infrastructure. Instead of working a deployment out on paper, you build it on screen, open source Standalone, Sentinel, or Cluster, or Redis Enterprise Active-Active and Active-Passive across regions, placed on bare metal, VMs, or Kubernetes, then route a key through it and run the failure modes to see how that exact layout behaves.

The infrastructure is the point. Nodes are fault domains, so where each master and replica lands decides what a node loss takes with it, and the deployment carries the rest of the picture too: what Redis holds, and the source of truth and caching pattern behind it. Routing runs the real CRC16 slot math, and the failure modes, Sentinel promotion, a shard loss, a region partition, memory eviction, hot keys, play out against the exact deployment you built. Everything is simulated, so you can take it apart without standing anything up.

Alongside it is a live flow simulator, the test setup the patterns come from: a two-node Kubernetes cluster running a six-node Redis Cluster across three shards plus Redis Sentinel, behind a FastAPI app. The builder is for planning a deployment; the flow simulator shows the same ideas running on live data.