▪ Designed and developed high-speed transmission systems for MiFID-compliance

Took a high-level architecture document and designed a software solution for the problem of copying 24-hours' worth of tick-data from all the stock-markets (up to 180Gb per day) and transmitting them to a remote location within 24 hours. This was a tactical, short-term product to help clients comply with the EU MiFID directive. Then went on to develop part of the long-term, strategic solution to the same problem which transmits the tick-data in 'almost-real-time'. Used Perl and C++ on Solaris, Windows and Linux.

▪ Planned, procured, provisioned and allocated test environments for web services

This was an interim management role between the unexpected departure of my predecessor and the hiring of a permanent replacement. I was responsible for 8 end-to-end test environments composed of Weblogic, Tibco, Gemini and several other software components. These had to be planned in detail, acquired, provisioned with software and data, then allocated to projects. Defects in the environments had to be analyzed and resolved or delegated. The configuration data for all the above was maintained in a spreadsheet. It was a very busy working environment with multiple projects competing for resources.

▪ Allocated, configured, prepared and maintained all non-production environments
▪ Legacy applications (off-the-shelf and bespoke) on both Sun and Wintel servers
▪ Migrated to service-oriented architecture based on BEA Aqualogic and Weblogic

This was a small, local firm that had grown extremely quickly (sales approaching £500,000,000 per annum). They engaged me to create the role of Environments Manager. I designed and implemented processes for creating, changing, booking out and maintaining 7 discrete software environments (for development and test). Each one comprised 20 servers (both Wintel and Unix) with multiple COTS packages and in-house developed applications (mainly based on the BEA Weblogic platform). They were starting to migrate all of this to a new enterprise service bus based on BEA Aqualogic. As the service-oriented architecture took shape, they were replacing existing hardware with Solaris virtual servers. I used BEA Aqualogic BPM Studio to model the relevant processes and created an Oracle configuration management database to maintain metadata surrounding the environments. As a result of my work, the Environments Manager

▪ is the single point of contact for all issues relating to non-production environments;
▪ is the first point of contact for resolution of non-production environmental issues;
▪ tracks and pursues non-production environmental issues through to resolution;
▪ liaises with Support, DBA and Delivery teams to ensure environment stability;
▪ proactively plans environment changes to support project requirements;
▪ is responsible for maintaining the schedule of non-production environments usage;
▪ has overall responsibility for non-production environment build and maintenance;
▪ schedules deployment of released code into non-production environments.

▪ Managed test and development environments, software builds and deployments

I assisted with the planning and management of both test and development environments, the performance of software builds and the deployment of built code into the environments. I used Windows 2000, Solaris 8, Korn Shell, Serena Changeman DS, Remedy and AttentiV.

▪ Developed a diagnostic reporting tool for users of HP's Ultrium (LTO) tape drives
▪ Reports identity, configuration, environment, health, performance & usage history
▪ Runs on several operating systems using embedded Perl within a C++ application

This was my fourth contract for HP (the others are listed below). Working with firmware engineers in the Bristol Peripherals Division, I analysed, designed and developed a diagnostic reporting tool for use by all owners of H.P. Ultrium (LTO) storage devices. The tool uses SCSI commands to examine device and media logs (held in EEPROMs), then reports on identity, configuration, environment, health, performance and usage history. It is available on several operating systems, launched as an embedded Perl object within the Library and Tape Tools application, which is available to all LTO users (written in C++).

▪ Corporate data-warehouse development project involving multiple legacy systems
▪ Responsible for the repository, test environments, workspaces, builds & releases

This is an on-going, corporate data-warehouse project in which I was engaged to assist with two major releases of functionality. The warehouse holds telephony data from a variety of sources and has links into most of the company's IT systems. My responsibility was the RCS code repository, with builds and deployments into developers’ workspaces, test environments and production. I maintained the Unix environments and co-ordinated their linking to legacy test environments for end-to-end integration and regression testing. I wrote shell-scripts for all the common SCM functions involving RCS and implemented controlled workflows to enforce activity-based change.

▪ Translated business schedules into Autosys 3.4 Job Information Language ("JIL")
▪ Developed a labour-saving Perl program to generate JIL from Excel spreadsheets

I performed software release and batch schedule management in a small team supporting around 100 development and test environments. Application components of several types were built and deployed and highly complex schedules for multiple Autosys instances were maintained, using Job Information Language (“JIL”). I inherited and further developed a JIL generation program (written in Perl). The tools used were Windows XP, Exceed, Lotus Notes, Solaris 8, Autosys 3.4, Perl and ClearCase 4.1.

▪ Reverse-engineered existing Java ETL programs, re-designed and re-coded in Perl
▪ Established the PVCS code repository, batch scheduling and monitoring capability
▪ Documented the application and trained IBM personnel to maintain and develop it

In this contract, the customer was NTL, which had outsourced its IT infrastructure and systems to IBM Global Services. The data warehouse ETL software had been written in Java and left undocumented by the developers. NTL then decided that it was a business critical application, so I was hired to re-design and re-write it. I created the development and test environments with a code repository (PVCS VM for Unix), reverse engineered the Java into a specification, then re-wrote all the programs in Perl and shell-scripts (including batch control). I documented the whole application, including maintenance processes, for IBM staff. During the last two months of the contract, I was asked to write a suite of Perl programs to replace Informatica ETL software in another data warehouse application, again documenting it all for IBM staff to support and maintain. One of these warehouses received its feeds from the NTL web servers, the other from interactive TV servers.

▪ Extensive IT systems development programme - Asset Servicing Transformation
▪ Wrote CM Plan covering OS/390 MVS mainframe, AIX distributed & PC streams
▪ Led 6-person CM team drawn from all 3 areas using Endevor, ClearCase & PVCS

This was a large software development programme (Asset Servicing Transformation) involving multiple platforms, technologies and environments. I initially wrote the CM Plans for OS/390 MVS mainframe, AIX distributed and desktop workstreams, then established a cross-discipline CM team. The core software was developed off-shore and I was responsible for the reception, configuration and deployment of code releases, which were tested in-house before going live. During development and testing, I performed the CM, build and release activities in the desktop workstream, using PowerBuilder with PVCS, WebSphere with Ant and ClearCase (writing Perl scripts to automate the processes). I was also involved in a company-wide CMM initiative, documenting the processes for configuration and release management. Tools used in the programme included OS/390 MVS, AIX, Windows NT, Endevor, ClearCase LT, PVCS VM 6.8, C++, Visual Age Java 3.5, PowerBuilder 6.0, WebSphere AD 4.0, Ant 1.5, Test Director 7.0i, Lotus Notes, HTML and Perl.

▪ Maintained ClearCase 4.0 & MultiSite installations in mobile telephony R&D labs
▪ With a colleague, supported 350 users on Solaris & NT, 12 servers and 100 VOBs

During this year I carried out deployment and support of ClearCase with MultiSite in a research and development environment for a leading mobile telecommunications equipment manufacturer. I and a colleague were responsible for 100+ replicated VOBs, 12 Unix servers (HP and Sun) and 350+ users (on Solaris and NT). We performed installation, upgrades, VOB creation/replication, Perl scripting, shell-scripting, job scheduling, configuration management, server monitoring, backups, disaster recovery, technical documentation, user support (out-of-hours included) and staff training. The tools used were Solaris 8, HP-UX 10.2, Windows NT, ClearCase 4.1, MultiSite 4.0, Korn Shell and Perl.