Michael Shaw Email & Phone Number

Christchurch, Canterbury, New Zealand

I worked as an AWS DevOps consultant applying the Well architected framework to client projects involving the use of AWS organizations, implementing organizational wide strategies for backups and patching to deliver enterprise level AWS account management.I also designed and deployed CloudFormation and CDK managed infrastructure to implement CMS sites, including CICD pipeline, following AWS best practices.Passed exams in AWS Advanced Networking, AWS DevOps Professional and AWS Security Specialty

Christchurch, Canterbury, New Zealand

I rejoined my previous team at MYOB on a short term contract to assist with an urgent migration - to plan and then migrate the PayGlobal Online service to Windows Server 2022. This was completed on time, with minimal customer disruption

Christchurch, Canterbury, New Zealand

I designed and implemented a performance testing client environment in AWS using terraform, Resources included the network (VPC/subnet/security groups/internet gateway) the compute power (autoscaling group/launch configuration/userdata script) the testing scripts (s3 bucket/s3 keys/iam role and embedded polices)I Identified gaps in the platform observability and monitoring. I participated in extending our cluster terraform pattern by implementing terraform resources to create datadog monitors.I reviewed our AWS infrastructure against company policies and risk register items and identified the areas that were not compliant. I presented solution options to the AWS account management strategy, and have designed a proposal for using a CI/CD pipeline based on terraform to create new accounts using the AWS Account Factory service managed by terraform.Provided engineering guidance and advice to information security team on several topics.

Christchurch, Canterbury, New Zealand

I identified the need, made the business case then planned and executed a change of FTP service to a commercial product that supported SFTP encrypted connections to enable secure transfer of bulk data.I implemented multiple CI/CD pipelines in Azure DevOps/Pipelines using a yaml/IaC approach to enable github version control on the pipeline definition. The implementation included testing, code quality and deployment steps via staging, and when authorised, into the production servers. The server deployment script was written in powershell.I created the company incidence response process, including the assessment of severity and impact of incidents, how the incident was communicated internally and the process around doing a post incident review to identify opportunities for improvement.

Christchurch, Canterbury, New Zealand

Passed AZ-204 Developing solutions for Microsoft Azure on the 9th November 2021.Designed a strategy to provide highly reliable access to HTTPS microservices. I implemented the healthcheck (is this instance healthy) and the service check ( can we service requests ) components for the microservice side, and defined the strategies for the circuit breaker and auto-retry client side mitigations to implement.Contributed to internal technical forums around architecture, testing and design of the overall service.

Auckland & Christchurch, New Zealand

Recruited from the UK to bring technical DevOps leadership to a new DevOps team that was inheriting the operation of a ClickOps managed platform that provided enterprise level payroll to NZ and AU enterprise customers (PayGlobal Online).Defined the technical vision for the team, Designed then lead the implementation of the CI/CD pipeline delivering AWS Infrastructure changes and Application software using AWS Code pipeline, Puppet and power-shell into a AWS EC2 network utilising Windows to deliver a Remote Desktop Application environment, Microsoft SQL Server, SFTP server and an IIS Web application.I planned and implemented Splunk log analysis strategy, redesigning the internal scripts from batch files into a standard toolset written in powershell with semi structured logging. Utilised complex event processing to alert based on combinations of events to improve integration reliability. Delivered the ISO27001:2013 compliance project for PayGlobal Online, wrote controls, identified defects, policy exemptions and presented evidence to the ISO auditor. Certified as compliant on the first attempt.Planned, scheduled and implemented many platform upgrades including AWS instance changes, software patching, and platform restructures - some requiring maintenance windows and customer notifications. participated in the weekly cross functional service meeting, ensuring all related parts of the business were aware of platform changes and we were continually adapting to our customers needs. Mentored my team, encouraged and developed their broader skills and recognised their successes. Embedded the pattern of automation and observably and the concept of self healing infrastructure.Subject matter expert for the DevOps skillset in the MYOB graduate program, mentored graduates through the program.Participated on the OnCall rota.

Harrogate, North Yorkshire, United Kingdom

IntechnologyWifi are a company that was attempting to create a commercial market out of providing free Wi-Fi in large public spaces. Their flagship deployment was in Edinburgh city centre.I started by analysing the current state of the infrastructure management, and identified a set of priorities for IntechnologyWifi to address.Phase I was to revolutionise their deployment of server side software through the use of automation and freely available tools, delivered the following:● A well defined Software delivery pipeline● TeamCity build server delivering software unit testing, artefact packaging with triggers for executinginfrastructure deployments when required.● The use of puppet to deliver predictable and repeatable deployments on EC2 instances.● The use of AWS Code deploy and bash scripts to ensure coordinated, reliable and reversibledeployments, including validation scripts that trigger automatic rollbacks when the validation tests fail.● Puppet components to implement the configuration of nginx, php and laravel third party software.● The use of AWS Launch Configurations and Autoscaling groups to ensure a robust and reliable platform for the software.Phase II was in progress when i resigned, providing the same automation to the deployment of AWS infrastructure changes.This has involved: ● Reviewing the current state of the AWS puppet automation modules available, and identifying that the key supported components have significant issues with their implementation. ● Created multiple ruby based puppet resource types using the type / provider model to managing the properties and lifecycle of some of the more challenging AWS components to model in puppet including launch configurations and auto scaling groups.

Huddersfield, United Kingdom

Control F1 are a company specialising in delivering commercial websites. Clients include some house hold names as well as a range of smaller businesses.I took on the devops lead role within Control F1, and has delivered on:● Application deployment for a multi tier telematics project with puppet / AWS Code deploy and AWSAutoscaling. Puppet facts (implemented in ruby) and puppet modules written in puppet scripts,templates and hiera following the Facts / Profile / Roles pattern.● Dynamic AWS IAM role and policy creation based on puppet facts and hiera configuration implementedin ruby as puppet resource modules● Continuous Integration and automated Deployment for multiple customer IIS deployments usingTeamCity, powershell scripts and web deploy packages.● Powershell scripted initialisation of new servers including software installation, configuration, warmup& verification and managed introduction into the active environment● Managing Control F1’s AWS infrastructure - build environments, shared servers and managed customer environments● Performance testing websites under developmentThere has also been several C# development tasks for a home e-retailer customer updating and extending their current website, and a complex data migration of products to their new CMS based website.

Clay Cross, England, United Kingdom

Formula E were building a compelling mobile application experience to complement their race series.Directed and assisted in the implementation of the server farm, from meeting with hardware suppliers, specifying kit, installing servers, deploying software, building an operations panel and monitoring the live system.Key details include:● Set up and managed the Development tool chain using Task management (Fogbugz), Source control (git), Build server (Team City), Repository (Artifactory). Used MAAS to deploy 40 servers, installed and configured riak (6 server cluster) rabbitMQ (3x 2 server clusters) as well as our own software and NginX used to handle load balancing across our internal HTTP services, and our public HTTP API.● Implemented real time reporting on the system behaviour using splunk to perform log analysis and generate operations panels and performance analysis.● Monitored the systems during the first live event, maintaining service throughout.● Specified Server specifications, cpu, memory, disk and I/O configuration, designed rack layout to maximise redundancy and optimise performance of the system as a whole.

Leeds, England, United Kingdom

Recruited as a developer on this role, but my biggest impact was in assisting the company towardsa devops style approach to monitoring and deployment. Their key products are ‘one off’ specialist welding machines designed to handle emergency repairs of oil & gas pipes at depth on the seabed. These complex machines are controlled by a net of integrated controllers.Key involvement:● Updating their development, testing and release process, using Kanban and git.● Introduced Isotek to logging and log analysis as a strategy to monitor a live system, allowing identification and investigation of unexpected behaviours in the live environment.● Updated the build and deployment strategy for the python packages, eliminating external dependencies on remote repositories, eliminating the need for development tools on client machines and ensuring developers were able to build and test the actual packages that will be deployed in the live system.

Leeds, England, United Kingdom

Recruited to act as a Technical Tester for performing non-functional testing on this project, howeverwith his developer experience, plus scripting and linux skills, his role quickly expanded into a software engineer role for the test team creating tools and scripts to enable the rest of the team to start performing their functional testing tasks with velocity.Key devops tasks performed include:● Deployment of software across 15 servers using puppet, including identifying and fixing puppet issues.● Creating operations panel for monitoring of test runs, reporting on performance, investigating data anomalies and making recommendations for areas needing more precise detail.● Scripting complex tasks into usable shell scripts to abstract away the complexities of an unfamiliar operating system and complex product.● Installation and configuration of Riak, RabbitMQ Redis and other software dependencies, updating puppet scripts to add additional configuration options and ensuring repeatability of the install and update process in the performance test environment.

Stoke-On-Trent, England, United Kingdom

bet365 is a UK based online gambling company with an international and substantial customer base that expect 24 hour website access.The primary projects were based around the NoSQL product GemFire. The data cache related projectsinvolved writing C++ and C# Gem Fire clients libraries providing GeoLocation data, capture tracking data and managing website session data. There was also a more complex task designing and prototyping a GemFire client library that provided transparent rerouting of GemFire requests to different GemFire server clusters to allow continuous website operation in circumstances that require a Gemfire cluster to be shutdown (e.g. Gemfire software upgrade).Http REST related projects involved investigating WCF’s 4.0 REST implementation, extending WCF in C# to incorporate Google Protocol Buffer serialisation into the channel, and identifying under stress testing the technologies unsuitability for bet365’s environment. Michael then designed and implemented a hybrid solution keeping the clean Service interface/channel approach used in WCF, and implementing a custom channel mechanism and using HttpWebRequest classes for the transport layer – allowing bet365 to keep the client ease of use of WCF whilst maintaining the HttpWebRequests ability to restrict the concurrent socket connections created on a client.Additional mini projects have involved the use of NoSQL databases Riak & Neo4j, improving the teams project release process to take advantage of AccuRev’s snapshot and crosslink features, and played a critical part in the rollout of a Gemfire C++ client library through the Website development teams and onto the production servers.

Watford, England, United Kingdom

Telehealth Solutions identified the need to reduce their fixed cost base to either gain further investment, or for sale to a new owner. To assist with this, I volunteered for redundancy and then continued to provide defined software functionality for a up front agreed prices.This line of work ceased when Telehealth Solutions was transferred to a new owner.

Telehealth Solutions is a provider of healthcare products aimed to enable patients to perform the routinehealth tests themselves, allowing the clinicians to centrally monitor patient’s wellbeing.I used a combination of C# with WinForms, XML, .Net Reflection and dynamically loading libraries tocreate a plugin architecture that enabled plug-in reuse across the companies entire product range.Typical day to day tasks included● Integrating medical devices, normally using C# communicating over RS232 serial ports to obtain measurements and other device status information. Occasionally C++ was needed to meet specific timing requirements, particularly common with non IrDA Infra-red connected devices.● Implementing and updating MS –SQL (2008) databases for the care-pod and occupationalhealth-pod products.● Connecting to Patient Management Systems. Typically using C# and COM, SQL or WebServices to talk to the clinical database.● User interface screens using C# & WinForms, including multi language support● User Interface screens using C# Xaml/WPF and multi language support.● Breaking work responsibilities into a list of actionable tasks, and presenting the tasks with enough detail that work could be shared amongst the development team.● Mentoring the junior developers.● Maintaining office network using OpenFiler SAN, vmware ESXi 4.0, Windows 2003 server Active Directory and an OpenBSD firewall.

Chesterfield, England, United Kingdom

At DSP Design, Michael was involved in writing C++ test programs to test and validate various Epic and PC104+ embedded computers. Tests development involved using a combination of either high level Win32 api's to play sound files, video files, echoing data over serial cables – these are typically production tests used in the manufacture and assembly process, and low level tests where bits and bytes are being written directly to chipsets validating hardware designs and correct functionality in the R&D phase.In addition to programming tasks, Michael crafted Windows XP Embedded images and Windows CE images for new processor boards, presented “hands on labs” at Microsoft events and became skilled in the use of a document system to investigate quality issues and following up with reports and recommendations to the board.

Brighton, England, United Kingdom

Raima UK was the UK reseller of the Velocis RDM. Michael used C++ to deliver new features on the company reporting tool including Unicode and MBCS support, added an enhanced user model to the Database engine, and created several wizards that assist the creation of reports using MFC. Michael changed the pre-existing company culture of “hacking code” and introduced source control & bug lists and the use of the C++ stl design patterns to the development team improving developer productivity.

Worthing, England, United Kingdom

Pocket Link is a small firm set up by a colleague from CSL. We created several pocket PC and Palm applications in C & C++. Michael and Barry tended work together using most aspects of extreme programming.

Guildford, England, United Kingdom

Argo Group was a venture capital funded expansion, delivering a “proxy server” software solution to improve customer experiences of WAP sites handling the limitations in the different device capabilities and enabling cross site portal branding. Development was performed following extreme programming methodology in C++ for Windows, Linux and Solaris servers. Used a lot of C++ stl design patterns.

Epsom, England, United Kingdom

Bacon & Woodrow is a firm of actuarial consultants.I Delivered a number of C++ ATL COM components that were used by internal customers. Products included a range of random number generators, distribution layers and an Implementation of Andrew Smith’s economic modeller.I also worked on the simulation of final salary pension funds project.

Guildford, England, United Kingdom

CSL, at that time was a data warehouse provider for the pharmaceutical markets. Michael Used Visual Basic to implement data manipulation in Access 95, 97 and SQL Server 7 databases and to extract the data in the right format for Esprit, CSL’s OLAQP data reporting tool. Michael spent his last year there working on the OLAP cube viewer, a C++ ActiveX control enabling n-dimension data analysis.