Brandon – C#, AWS, Azure DevOps

The data and analytics layer of a smart-building platform. It pulls device telemetry and operational data from several databases into one place for company reporting and near-real-time operations dashboards. ML models run on top of it to detect device anomalies, score energy and operational efficiency, and identify devices automatically across large operational datasets.

• Architected and delivered production ML models for anomaly detection, efficiency scoring and automatic device identification.
• Built a cross-database dashboard as a stopgap reporting solution, then designed its replacement on Microsoft Fabric Lakehouse with PySpark and Power BI. The replacement was partly delivered when the company was sold.

A SaaS smart-living platform for condominium developers and property managers. Residents use a wall-mounted panel and a mobile app to control devices in their suite: HVAC, lighting, blinds, leak sensors, digital door locks and Stay/Away security modes. At the building level it connects to elevators, access control, smart lockers, amenity booking, licence-plate parking and energy monitoring. Each building runs its own edge server for 200–2,000 suites, connected to a cloud backend built to handle 10,000+ suites.

• Architected and led the build of an in-house smart-living platform that replaced a closed third-party product, where even small changes took about a year. This gave the company full control of its roadmap and an easy way to add new third-party integrations.
• Designed the event-driven, multi-cloud architecture: an edge server in each building, Azure IoT Hub to bring device data in, Event Grid with retries and dead-letter queues, auto-scaling Azure Functions, and AWS API Gateway/ECS services. It was sized for 10,000 suites with about 10x headroom.
• Turned an open-ended "replace it" brief into concrete requirements with business stakeholders. Changed resident onboarding from remote invites to in-suite QR scanning so residents had to be physically present, which met the security requirements.
• Built the elevator-integration microservice. When the first assumptions didn’t hold, redesigned it to handle complex setups such as stacked elevators.
• Led the company’s DevOps adoption: CI/CD pipelines, automated deployments, and infrastructure as code with Terraform and CloudFormation. Trained a network engineer into the DevOps role.
• Led a team of up to 8 backend and DevOps engineers, and ran hiring loops, technical interviews, code reviews and architecture mentoring.

A hiring platform for employers who can’t tell from a resume alone whether a candidate will fit their existing team. Applicants send in video applications. A third-party personality-compatibility engine compares each applicant with the current team’s profiles to judge cultural and interpersonal fit. Employers see assessments, behavioural insights and videos side by side. The platform serves several employer organisations from one high-volume system.

• Designed and built the .NET Core back-end for the multi-employer platform, including a secure integration with the personality-compatibility service.
• Built API integrations and automated pipelines that brought candidate data in automatically, removing manual data entry.
• Led a team of 2–8 engineers and worked with stakeholders to set requirements and engineering priorities.
• Built internal diagnostics and developer tools that made troubleshooting faster and new engineers quicker to onboard.
• Mentored junior and mid-level engineers, which made delivery more consistent and prepared them for product engineering roles.

A SaaS compliance platform that US employers use to produce IRS Form 1095-C under the Affordable Care Act (ACA). It applies complex ACA eligibility and coverage rules to large volumes of employee data and produces accurate reports every year, which lowers employers’ regulatory and penalty risk. It was the company’s main reporting product and also covered onboarding new employers.

• Redesigned critical batch-processing workflows, cutting a monthly data process from 52 days to under an hour.
• Re-architected the data infrastructure, merging 100+ SQL Server instances into about 10 with no loss of performance and lower AWS costs.
• Led database tuning and caching for employer onboarding, cutting page load times from over 2 minutes to under 3 seconds.
• Went from Junior Developer to leading a team of 7 engineers within one year, while still writing code.