Public sector & enterprise · 2013–18
Mission-critical platforms
Five engagements for the NHS, government and energy, where the work was hosting, assurance and sign-off.
- Role
- Developer → architect & technical lead
- Sector
- Government · Healthcare · Energy · Events
- Period
- 2013 – 2018
- Engagement
- Five engagements
- Worked with
- WordPress, AWS, AWS PrivateLink
Some clients don’t play by the usual rules of web development. A government department that can’t discuss where its website is hosted. A health service where downtime hurts people. An oil company whose compliance requirements arrive before the brief does. A race venue with one weekend a year that matters.
Between 2013 and 2018 I kept ending up in these rooms. The engineering was rarely the hard part. What these clients needed was evidence, process, and someone willing to be accountable.
What the work involved
A sensitive government department. On paper, the FCO Services website (now FCDO Services) was a simple WordPress build. In practice the CMS barely mattered. The work was in the hosting, the security, and the assurance a sensitive department needs before anything goes live. My first lesson in the gap between “it works” and “it’s approved”.
The NHS. Geni, the platform behind the NHS graduate management training scheme. Not glamorous, but built to be dependable for an organisation where “it broke” isn’t an acceptable answer.
A government department, no code at all. UK Trade & Investment asked me whether to renew the technology partner behind a core system or pick a new one. I reviewed the options, costs and risks so the department could make the call. Sometimes the most useful thing you can deliver is a document.
Shell. Ideas 360, a global platform for ideas and collaboration, built on the Twine product with custom components for Shell. Meeting Shell’s compliance and documentation standards was most of the work, and it’s still the bar I use for “enterprise-ready”.
A world-famous motorsport venue. A multi-lingual events website with ticket sales through AudienceView, and live text and voice chat wired into an on-site Avaya phone system over AWS PrivateLink. For a few days a year the whole world turns up, and nothing can fail.
None of these clients could afford a second attempt.
Where it stands
A lot of my working habits come from these projects. Infrastructure as code, so environments can be rebuilt and checked rather than described. Audit trails and access controls built in from the start. Documentation written for the people who approve a system, not just the people who maintain it. And asking “who’s accountable for this decision?” before “how do we build it?”.
I bring the same habits to everything else, including where nobody strictly requires them.