Andrew McGrath

I joined Epic Games (child company SuperAwesome) as a Senior Platform Engineer and worked in a team focusing on an Epic Games platform project. This was to develop an extension to an existing in-house Kubernetes cluster lifecycle management service to provide update scheduling and automate customer communication via Slack.

The project was written in Go with an event driven architecture and provided the the following features:

• Allow scheduling windows to be created by the relevant Platform Team members
• Identify Kubernetes clusters that are in an upgrade ready state
• Automatically assign upgrade ready Kubernetes clusters to upgrade windows
• Notify Kubernetes cluster owners of upgrade scheduling events via Slack.
• Identify health checks that require customer action to resolve and automatically notify them via Slack

I moved to become the Technical Lead one of the Platform teams initially in SuperAwesome then directly with Epic Games. While working as a Technical Lead:

• Led a high-performing team of four engineers supporting SuperAwesome and Epic Games development teams.
• Implemented agile practices to optimise workflow and improve efficiency within the DevEx team.
• Worked on implemented Go as the primary programming language for the SuperAwesome platform team. Building up standards, processes and a set of commonly used packages.
• Worked to migrate Helm charts, NPM and container images to be stored centrally on Epic's internally hosted JFrog Artifactory.

While here Epic Games made the decision to split and sell part of SuperAwesome. This lead to my team focusing on migration work to separate/split the platform for SuperAwesome and the service Kids Web Services (KWS), which was to be retained by Epic Games.

Joined as part of a newly formed Cloud Centre of Excellence team focused on providing cloud resources to the broader Platform teams, software engineers, and business.

The teams and my responsibilities:

• Provide core cloud resources utilising infrastructure as code (IaC). The team's preference is to use Terraform and Terragrunt to provide this.
• Ensure cloud resources adhere to security and company policies.

I also worked on the following improvements:

• I aided with the migration from GitHub and GoCD pipelines to GitLab and GitLab CI pipelines.
• Led delivery design discussions and helped introduce IaC unit and integration testing with Terratest.
• Work with the Engineering Manager and the team to move towards Agile, Continuous Improvement, initially with the adoption of Scrum.

Was the second Platform Team member and help to design and develop the Internal Developer Platform and infrastructure required for the business needs.

• Built an Internal Developer Platform comprising of guided and self service solutions for the wider engineering team. This prompted the DevOps ethos "You build it, you run it" for the technical side of the business.
• Promoted Continuous Improvement within the Platform Team driving work flow and process improvements to increase productivity and focus on the value delivery.
• Worked to build and refine a technical recruitment process for Platform Engineers including interviewing, technical reviews and on-boarding/mentoring.
• Provided solutions for the platform, development experience and deployment with the use of Infrastructure as Code (Terraform), CI/CD pipelines, service templates (.Net templates) and deployment templates (Helm). Additionally designed and built the business infrastructure on Azure, Kubernetes, with Linkerd service mesh, Traefik Ingress controllers, Cloudflare CDN/Security with a dedicated internally build observability solutions around Prometheus, Grafana, OpenTelemetry, Loki and Tempo.
  

The following is some of the soultions and associated technologies used within this position:

• Running and supporting multiple Kubernetes clusters, running on Azure (AKS) automatically provisioned with Terraform, Helm and CI/CD pipelines.
• Advocating for the use of and implemented a Service Mesh (Linkerd) within Kubernetes.
• Automated monitoring solutions, auto-creation of Grafana dashboards, panels and alerts.
• Cloudflare (Enterprise) as the CDN and WAF/Security frontend, managed via Terraform (where possible).
• Azure DevOps including CI/CD Azure Pipelines, developed as a set of reusable templates.
• Built a Distributed Tracing solution including writing and maintaining an internal frontend (JavaScript) instrumentation module to work with Next.js.
• Wrote a Kubernetes Event Scraper service in Go to listen for and filter Kubernetes events and create Grafana dashboard annotations.
• Worked closely with the Engineering teams to improve development experience including CI/CD pipelines and local development environments.
• Assist with security both for the services and internal staff.

Co-Founded Strata Labs, a specialist F5 consultancy with a focus on IaC and delivery automation.

• Heavy use of GitLab to build CI/CD pipelines, using Docker containers for Ansible deployments and testing using Postman/Newman workflows.
• Automated the migration of F5 services for international utility provider Engie, by building Ansible Roles for various services.
• Using Ansible roles alongside Jinga2 templates ensured consistency across environments in an effort to limit configuration drift and speed up delivery

Initially worked as part of a project-based agile development team with the primary focus on developing F5 iRule (TCL) and iRuleLX (Node.js) solutions.

I then moved to work as the Lead Software Engineer on a greenfield internal tooling project to provide a self-service ability for Value Streams needing to configure Akamai for cloud-based security and DDoS protection. I helped design and build the solution as an API application with a declaration model. It allowed for a reduction in Akamai initial service configuration time from 8-12 weeks (via a manual configuration team) to less than an hour and changes to be part of CI/CD pipelines.

Key Projects

• Cloud Protection Automation - The solution is developed in Node.js using Express and runs in an Azure Kubernetes Services (AKS) with Istio service mesh and utilises Azure Key Vault, Azure Cosmos DB (Mongo) and Azure Active Directory for authentication via OAuth. As the lead engineer, I managed a team of three other engineers, defined workflows and standards, worked with project and business teams to define and prioritise feature development.
• F5 iRule Optimisation - A critical project was to identify and reduce the CPU utilisation of iRules by optimising the code. Here I wrote several small tools to enable the capturing of iRule resource utilisation, identifying the highest resource utilisation iRules. The project achieved an estimated 20% total decrease in CPU utilisation by the iRules allowing for an increase in user capacity in a vital area of the bank.
• Horizontal Scaling F5 - Project was to increase the number of edge F5 BIG-IP system by scaling horizontally. A redesign of the iRule DC Affinity solution was an essential requirement along with infrastructure changes and the updating of multiple tools for deployment, monitoring and system management. The outcome of this was the near doubling of end-user session capacity within the targeted tier of the bank.

Key Achievements

• Building and improving several CI/CD pipelines in Azure DevOps, Jenkins and UrbanCode
• Automating parts of the infrastructure with the use of Python, Jenkins and UrbanCode
• Expanded the development capacity of the F5 team with iRuleLX (Node.js)
• Development of Node.js/Typescript framework for iRuleLX including an internal F5 Node.js RPC local test client/server
• Improved the in-house unit test framework, written in TCL.
  

F5 Consulting for a set of large scale infrastructure projects migrating and replacing F5 systems. The projects were made complex due to a large number of configuration migrations required, and it was a common requirement for multiple F5 configurations to be merging into a single device configuration.

I helped develop several of the tools used to automate part of the process, enabling a higher pace of migrations that would have otherwise been possible.

I finished this project performing a critical F5 GTM migration which involved a large number of network and firewall changes and was a critical dependency to further F5 configuration migrations.

As part of HP Enterprise Security Services (HP ESS), a security division of HP Enterprise Services, I worked designing and implementing enterprise-class security solutions for HP's customers with a primary focus on F5 Network solutions.

Key Projects:

• Tesco Bank - Cisco to F5 migration. I worked as F5 technical lead designing and building a new F5 vCMP environment and underlying network within Tesco Bank and migrating the existing Cisco ACE and GSS configuration to F5 Silverline, LTM and GTM.
• AXA Insurance - Cisco to F5 migration. Leading a small technical team and developed in Python a configuration conversion tool which allowed me and the team to migrate over 3000 VIPs from Cisco ACE to F5 LTM across several environments in just over 3 months.

PCI DSS project to migrate two key application to the Cloud, I designed and implemented a Windows Application solution in AWS cloud to ensure compliance with PCI DSS.

• Senior Security Consultant (F5/Juniper) - NTT Comms Security
  
• Security Consultant (F5/Juniper) - HP Enterprise Security Services (now DXC)
  
• Technical Specialist (Networking/Security/Storage) - Acora
  
• Technical Consultant and Certified Juniper Instructor (Juniper) - Sphinx (now Arrow ECS)
  
• Senior Systems Engineer - ADA Computer Systems (now Acora)
  
• Systems Engineer - Vantage Computer Systems