CV
I’ve worked in the financial services industry most of my career, with companies such as Goldman Sachs, Credit Suisse and more recently Man Group. Putting it simply, I specialise in getting stuff done. I sit right in the middle of a bunch of agile
development teams, and I help them to get their own stuff done quicker and better than they could if I wasn’t there.
If you’re not into reading long resumes, read the following few bullets for a whistle-stop tour of what I’m good at:
- Linux (RHEL, Ubuntu) and Windows Server administration.
- Deployment and configuration management with Ansible.
- Scripting with Bash, Powershell, Batch.
- Development with Ruby, Ruby on Rails. Hacking on anything else thrown my way.
- Monitoring servers and applications with tools such as InfluxDB, Telegraf, Chrongraf, Kapacitor, Sensu, Collectd, Grafana, Splunk, Monit, Consul, Nagios.
- CI with TeamCity, version control with Git, SVN, Bitbucket Server, Github Enterprise.
- Multi-region team management, including hiring, performance reviews, objective setting, workforce reduction
You can download the full version here.
Employment History
Man Group - September 2016 - Present
DevOps Engineer
Specialising in all things that make developers lives easier:
- Deployment of the InfluxData “TICK” stack - Telegraf, InfluxDB, Chronograf and Kapacitor along with Grafana - allows developers and support teams insight into the state of infrastructure and the applications running upon them. Worked with lead developers to assist them in creating dashboards useful to the development teams.
- Development and deployment of a ChatOps bot (based on GitHub’s Hubot), giving developers easy access to the deployment status of their applications from within HipChat chat rooms. The bot interfaced directly with an Octopus deployment server, as well as queried InfluxDB for application service status. It was also hooked into the api.ai natural language processing service, allowing the processing of natural language queries for bot functionality.
- Designed the deployment architecture for deployment to Linux boxes (with no root access) from the existing deployment solution (Octopus), as well as via Ansible.
- Drove open source contribution by creating the Man DevOps GitHub page, publishing development that’d be useful to the community.
- Devised an automated software versioning and deployment strategy utilising Git, GitFlow, TeamCity and GitVersion. Worked with development teams to migrate from SVN and onto the solution. Assisted developers with migration from SVN to Git.
- Build of high-availability monitoring and alerting stack with Sensu. Alerts funnelled through to Alerta for de-duplication before being delivered to stakeholders ChatOps-style into HipChat rooms.
Credit Suisse - January 2015 - September 2016
I ran DevOps (i.e. “Getting Stuff Done”) for the Equities Risk division.
- Management of a 1000+ server estate consisting of both physical and virtual servers, split approximately 85% Linux,
15% Windows.
- Configuration management of infrastructure and deployment of software using Ansible.
- Physically located within an Agile dev team - involved in multiple daily team standups, production issues,
iteration planning. Work closely with devs and operations teams to attempt to break down traditional banking “siloed”
roles.
- Monitoring and capacity analysis using Collectd, Graphite, Grafana, Monit, Nagios. Built relevant dashboards to
monitor JVM & HTTP based applications. Understand the intricacies between monitoring VMs and physical hardware.
- Maintenance of deployment scripts in Ruby, deploying to both Linux and Windows hosts.
- Performance tweaking of Windows and Linux infrastructure to maximise throughput for JVM & HTTP based applications.
Some of the larger projects in this role:
- Managed and executed replacement of 600 virtual RHEL Linux servers and application migration to 150 physical
servers, for purposes of cost reduction, reliability and performance consistency. Post provisioning configuration and consistency checks were carried out via Ansible. Servers were split for use by multiple services using c-groups.
- Development of automated deployment tool for dev, UAT and production environments using Teamcity and Ansible.
Allows new services to be deployed and released simply by developers including an Ansible playbook within a build package.
- Development of a service “baseline”, consisting of a repeatable set of tools for use by new services; Ansible
for deployment, Collectd and Graphite/Grafana for metrics, Splunk for log collection, Nagios/Monit for monitoring
and Consul for service discovery.
Credit Suisse - May 2011 - December 2014
AVP, Environment & Release Management, Equity Derivatives
I managed the Release and Environment Management team at Credit Suisse, a team of eight, split across multiple
regions. The team is responsible for all software released into the Equity Derivatives division, the infrastructure
that it runs on, and the infrastructure used to develop it.
My role was extremely varied, but not limited to:
- Capacity and platform management - provisioning, decommissioning, capacity analysis. I lead the project to
introduce Nagios into Credit Suisse, which now monitors the entire Equity Derivatives estate.
- Technical consultancy - I worked closely with development teams to help them to select the best technology
and infrastructure for their application. This often involved introducing new previously-unapproved technology
into the bank and arranging approval for use.
- Strategic software development to assist my day to day role, e.g. scripting large scale software deployments
across multiple platforms.
- Release management - deploying the software produced by various development teams with Equity Derivatives IT.
I ran the team borrowing practises from common agile development methodologies. Rather than a weekly team meeting,
we had daily stand-ups and monthly retrospectives. With the high turnover of work, this allowed me to keep on top
of the teams’ upcoming deliverables ensuring targets were met.
Credit Suisse – July 2009 – May 2011
Working within a small team, I was responsible for an estate of over 3000 mission-critical servers used to
calculate intra-day and nightly risk for all key business units within the bank.
- Day to day management of a 3500 server estate, utilising custom-written scripts to assist in management and
fire-fighting.
- Implemented Splunk for capturing of logs and performance metrics of servers in the grid.
- Wrote a custom Powershell module to assist team members in managing grid nodes.
- Deployed Platform Symphony to replace in-house written grid scheduling software.
Goldman Sachs – January 2006 – May 2009
System Analyst
Working within a team of five I was responsible for 3rd line support of all Windows Server infrastructure within
Goldman Sachs EMEA.
- Maintaining approx 5500 Windows XP workstations, 1000 Windows 2003 mission critical servers, including patching,
performance monitoring (using custom perfmon scripts) and application deployment.
- Project managed two new office setups from a Windows Infrastructure perspective. Included designing workstation
builds and writing group policies to suit.
Alea Group PLC – June 2005 - January 2006
Technical Analyst
Working within a team of five I was responsible for 3rd line support of a 70 server and 160 user Windows 2000 / 2003
Active Directory network, and was technical lead on a large proportion of Alea’s system project work.
Rank Group PLC – December 2003 - June 2005
Technical Analyst
Working within a team of six Support Analysts, I was responsible for the 2nd and 3rd line support and maintenance of
a 5000 user, 160 site Windows 2003 Active Directory network.
VEGA Group PLC – June 2003 - December 2003
Technical Analyst
My first job after leaving university, working within a team of three I was responsible for the computer support of
50 on-site users as well as approximately 150 mobile users.
Education
University of Kent at Canterbury - October 1999 – May 2003
Multimedia Technology and Design - First Class with Honours
Sandhurst School - September 1992 – September 1999
3 A-Levels, 10 GCSEs
