Matthew Macdonald

Led the migration of a critical application from legacy infrastructure to Azure + Kubernetes, improving scalability, resilience, and long-term maintainability

Re-architected application code (.NET) from an Azure Function into three Kubernetes-native workloads (API, Worker, CLI), enabling horizontal scaling and cleaner separation of concerns

Migrated CI/CD pipelines from Azure DevOps to GitHub Actions and Octopus Deploy, aligning with newly established organizational engineering standards

Automated the data migration, minimizing downtime and reducing risk during cutover

Authored and executed the cutover strategy, ensuring seamless transition of the application with minimal disruption

Implemented IaC to automate database provisioning within Azure landing zones, enforcing least-privilege access controls to databases hosted on a shared SQL Server and reducing manual effort across environments

Uplifted and standardized shared Terraform modules, accelerating adoption of cloud best practices across teams

Built an internal .NET Blazor web app which allows platform team members and devs to quickly and efficiently manage the secrets in their environments

Tasked with bringing a collection of Azure infrastructure supporting the sales pipeline environments up to platform standards. This requires a deep dive into the existing infrastructure and config (undocumented) to gain a strong understanding of the landscape before recreating it from the ground up using Terraform, while documenting everything along the way and creating playbooks for prod deployment

Work with dev teams to gather infrastructure requirements and advise them on correct patterns and best practices

Contribute improvements to the internal Terraform module library

Reduced total monthly Azure spend by 40% by consolidating app services into modern plans across all environments and other miscellaneous changes

SME for Kolmeo’s self-serve environment platform, built on Kubernetes using Terraform, Helm & scripts. Ongoing maintenance, support, and improvement of the platform

Instrumented OpenTelemetry within our application – providing dev teams with the ability create vendor agnostic custom traces and metrics

Modularised our platform layer, enabling multiple versions of it with different use cases and security domains. This allowed rapid creation of dedicated platforms for things like trials/demos and migration dry runs which improved our sales and onboarding pipeline

Created and maintained an internal web app (Node / Vue.js) for the execution of approved SQL scripts against production DB’s – reducing load on data & operations teams

Using C#/.NET, added MFA config and auditing functionality into our main application. This is an example of real production feature development, including unit tests

Performance tuning and optimisation of our platform infrastructure, comprising of memory optimised nodes, spot pools, SQL resource pooling, workload scheduling and node autoscaling. This reduced both env creation time and cost by 87% each, coincidently

Redesigned the infrastructure of an Azure hosted web application, taking it from a point-and-click delivery style to a GitOps structure defined in Terraform, allowing near instant DR

Overhauled CI/CD pipelines, increasing efficiency and reliability through dependency caching, agent runtime version locking and job concurrency, which increased developer confidence, enabling a higher release cadence

Implemented dependency vulnerability scanning on pipelines, illuminating multiple critical CVE’s, substantially reducing risk and securing the SDLC moving forward through quality gates

Identified and implemented cloud resource cost reductions, reducing the bill on cost-majority resources by 25%

Member of the Information Security Working Group, responsible for the implementation and ongoing maintenance of Business Fitness’ ISMS, fulfilling requirements of ISO27001

Instrumental in the diagnosis of multiple production issues, in one case implementing APM tools on the fly to diagnose a buggy message queue publishing method. In another case utilizing Grafana and dotnet debugging tools to discover a thread starvation issue on a Kubernetes deployment that was causing heavy disruption

Research and development of solutions to new and existing problems across Dean & Co’s client base, consulting through the process

Successfully led and delivered multiple projects across multiple sites

Implementation and administration of the cloud tenants across Dean & Co’s client base

Architected a solution in Azure with the requirement of a Microsoft SQL server being accessed by IoT devices throughout their sites, inclusive of security best practices

Implemented an automated device imaging solution, decreasing manual labour by 50% and increasing endpoint reliability by 25%