IT Consultant / Software Architecture / Dev-Ops / Development / Cloud / Monitoring

Part of the SIEM team at VW Financial Services that engineered and implemented a large QRADAR deployment from the ground up.

Was part of the team developing new external support tooling and systems to integrate with the core QRADAR SIEM system to deliver new quality control and software improvement processes specifically to provide continuous versioning of parts of the user data model within QRADAR.

I designed and implemented a Domain Specific Configuration Language (DSL) loosely based on the Hashicorp HCL2 reference and a parser for this in GoLang in order to allow seamless querying, versioning, synchronization and adjustment of QRADAR configuration items (specifically Custom Event Properties and associated dependent types and subtypes) to and from QRADAR and an externally versioned persistence layer using the QRADAR JSON based RESTAPI interface. I also setup a proof of concept environment consisting of a QRADAR instance and a GitLab instance along with a CI / CD pipeline in order to validate build and execution sequences.

Architecture, Design & Implementation of a Proof of Concept for a Cloud Based Monitoring Sub-System to display, in near-real-time, continuous transaction conformance to Business SLA's for a private cloud based application with external service dependencies.

The project required analysis & integration of various monitoring technologies (e.g. Influx / Cassandra / Prometheus / TimeScale). Each of the technologies was integrated using Java and SCALA API's into a sample version of the client's SCALA application along with custom dashboards. The pro's & con's of each technology along with sample data from dashboards were presented to the client, and based on a final presentation to the client, a customized solution utilizing TimeScale was arrived at, (based on ease of maintenance and standard SQL support from the Postgres support beneath TimeScale) and then further JAVA & SCALA code refinements were integrated into the clients' final application, and deployed using scalable Docker containers into a private cloud.

Worked within Agile teams to develop new log and event exporters for application level metrics ( Functional & Non-Functional ) as part of the migration of the in-house monitoring system ( Sentinel ), and served as part of a larger Proof-of-Concept to develop a cloud based Monitoring-as-a-Service ( M-a-a-S ) concept based on a microservices architecture utilising  Prometheus, Thanos, and Grafana all running in Kubernetes and OpenShift in an On-Premise private cloud ( ARGOS ). I also looked at alternative Proof-of-Concepts that were lighter weight / lower bandwidth solutions for data centre scale data collection using alternative architectures and technologies such as local node data collection with FluentBit and centralised log processing with FluentD inside Kubernetes and OpenShift based systems.

• Migrating legacy Python code base Metric Exporters into GoLang as well as writing new Custom Metric Exporters natively in GoLang.
• The custom exporters utilised the HTTPS REST interface of SPLUNK to extract the relevant data and used the Prometheus GoLang client libraries to present the metric data in a Prometheus scrape-able manner.
• Unit tests were created in the Golang code and once successful testing was complete via the CI/CD pipeline, the exporters were deployed into Docker Containers within predefined Pod’s and run in an OpenShift environment.
• Sample Grafana Dashboards within the OpenShift environment were created to test that the Prometheus metrics were valid
• Providing assistance for Cloud Migration to other Amadeus Public Cloud projects targeting AWS & Google Cloud ( S-Box / X-Box ) by migrating the ACS project scripts using Terraform, Ansible etc
• Proof-of-Concept monitoring with FluentBit and FluentD

Implemented new functionality in the Airline Dashboard Project within an Agile team developing custom in-house ETL ( implementing data collectors in Python to extract, aggregate, clean & transform data ) to provide sanitised data views to Business Intelligence teams in a customisable dashboard for end-customer airlines such as AirFrance & Lufthansa etc. Also provided Cloud Support for Cloud Migration Projects based on the reference Amadeus Cloud Services architecture.

• Developed new Docker services using Python and GoLang to interface to legacy Flight Booking Systems via REST API’s ( micro-batching techniques to extract in near real-time ), to extract details from Passenger Name Record ( PNR ) data such as Airline ID’s & 3^rd party office ID’s and re-aggregated these to generate new time series e2e monitoring views of systems accessible via REST & Grafana.

• Developed custom metric exporters in Python which interfaced to BI Systems which used ELK and executed Elastic Queries
• The results of the Elastic queries were aggregated and reformatted and exposed via REST as well as made Prometheus scrape-able.
• Created sample dashboards in Grafana to test that the Prometheus metrics were available.
• During the project, ELK was phased out and migrated to SPLUNK, so the project was migrated, re-written & adapted to target SPLUNK.
• Providing assistance for Cloud Migration to other Amadeus Private Cloud projects targeting RedHat Private Cloud ( R-Box ) by migrating the ACS project scripts using Terraform, Ansible etc

Architecting and implementing a microservices based cloud monitoring solution for an e-Commerce business unit within the Amadeus Cloud Services (ACS) private cloud for the back-end e-commerce booking engine for the Inter-Continental Hotel Group (IHG) booking website. (The proposed high level solution was based on the Netflix Lambda Architecture.)

I worked in an Agile team of 5 developers in a DevOps environment, utilizing the following technologies for the solution: Java based Middleware, Kafka Broker Cluster, Apache Spark Cluster, Zookeeper, Diamond, InfluxDB, Grafana, MapR, and FluentD along with Docker, Puppet & Hiera, to implement a proof of concept in Openstack (VIO + RedHat OpenStack), and RedHat Linux on Openshift.

• Implemented a CI / CD nighty build system based on an implementation of a Jenkins server with custom workflows to pull BitBucket source code from each of the repos, build the resulting artifacts into Docker images & then test, tag and push the validated images to Artifactory

• Created an on-demand workflow which would create the MVP cluster of VM's needed within the openstack environment and populate with the latest versions of docker images from Artifactory to test environment viability at that point in time.

• Coding and maintaining parts of the monitoring workflow using Python, JAVA, and DockerFiles and associated build tools such as Maven and Makefiles, to build and wrap components such as Diamond / FluentD / Kafka / Influx / Grafana and IAM / LDAP clients in a Lambda Architecture along with unit tests in a local environment, packaging, deploying and running. ( development to production )
• Extended & Created new Puppet & Hiera definitions for VM’s for final environment specifying custom definitions for logging subsystems.
• Developed custom in-house tooling ( ULF-CLI ) in SCALA as SPARK client within a MapR HADOOP cluster to allow rapid searching for application events within the AVRO encoded files within the MapR HDFS repository, given various regex & time patterns.
• Developed in SCALA an automated cleaning tool to maintain the optimal amount of data in the HDFS before triggering HDFS clean up.
• Working with QA teams, application teams and SRE teams to test and deploy updates and patches across various environments.
• Providing real time assistance and support to 1 ^st and 2 ^nd level support teams with the monitoring system after Go-Live.

Technical Application Manager - Responsible for a range of Agile projects within the Consumer Channels dept of Vodafone GA :

• Providing technical support at all layers of the application SW & HW stack for the entire project portfolio for Vodafone Business Units throughout the various key project delivery lifecycle stages from Architecture Design Phase / Environment Rollout / Production & post Go-Live and SLA definition & implementation along with SLA Reporting & Dashboarding.

• Design & initial implementation of a roadmap for a fast-track training & on-boarding plan for a global dev-ops team resulting in a ‘follow-the-sun’ model across all near- and off- shore teams in order to reduce costs across the dept.

Worked within a small Agile Dev-Ops team to architect, design and implement an automated ‘on-demand’ environment rollout for a German Government project for Pharmaceutical Evaluation ( AMNOG ). The solution was based on a highly secure version of the Document Management Stack from Oracle ( i.e. Apache / Weblogic / Web Center / Web Center Portal, Oracle RDBMS, LDAP & IAM ) which required end-to-end TLS security concepts and implementation across all layers of the application stack. I worked on this project from the original design through to final implementation & Go-Live in Production in a data centre in Schwerin, as well as providing post Go-Live support.

Deployment & Rollout of the SEPAHUB financial payments clearing system across all Vodafone operating companies across Europe. Including multi-environment setup ( DEV / TEST / FAT / Pre-PRD / PRD, etc ), OS install & configuration across all RHEL Linux VM’s; Config mgmt & automation of packaged software management with Puppet & RH Satellite Server; definition, creation, automation of package checks with Ansible as well as end-to-end smoke testing etc. and provision of 3rd Level Support to the SEPAHUB Production Operations teams.

• Definition and implementation of a roadmap for a new deployment process in the project to implement a CI/CD approach for all rollouts.
• Troubleshooting performance and stability related application and system issues during Acceptance Testing phases.
• Deployment Support, ( to teams in Germany & India ) for ATG Commerce, Oracle, WebLogic Middleware, Solaris, Apache, and IAM & LDAP systems.
• Architectural Design support for the proposed SEPA pan-European bank integration in 2014 of the Vodafone Germany Online Shop project
• Solaris and RHEL based System Administration for various test and prod environments.

Live ! : Operations Team Lead & Single Point of Contact for Deployment & Integration activities - in charge of a team of 5 Agile specialists responsible for maintaining the core application stacks to provide 24x7 service for 5 different Vodafone Live! country portals and managing the client interface with the relevant opco's for each country portal as well as analyzing and implementing performance and stability enhancements for all components of the Live! portal architecture on the production systems.

MyWeb: Team Lead within the Mobile Portals team. This role involved leading a small team of 5 Agile specialists and solution architects for the MyWeb (both web and wap) portals to analyze, isolate and optimize key performance blockage points within the distributed multi tier middleware based architecture and to alter the configuration and deployment procedures for future releases to account for and to update and propagate any changes made via the SVN systems in place.

This role also involved providing migration & configuration support for the Vodafone Live! and MyWeb teams to assist in the migration of supported devices and users from the Vodafone Live! Portal to the MyWeb portal.

Core Enablers Team: Providing configuration, deployment, integration and debugging support for apache and complex java based web applications in the production and pre-production ER systems (Vodafone Global billing and mediation systems) for all opcos.