1,000 Days

One thousand days is a long time.

Life has changed immeasurably for everyone on the planet. As some of you know, I decided to go for the grand slam of life changing events as a global pandemic clearly wasn't enough for me. For once, I'm going to take the path of least resistance and list what hasn't changed:

• I work at Form3.

I'm not going to sit here and tell you it's because everything is perfect here - it's absolutely not - but a wise woman once told me, "it doesn't have to be perfect, son, it just has to be worth it." Hopefully there's something useful in this article whether you're going through similar turbulence or simply need something to read with your coffee.

Day Zero

The most interesting thing about my arrival at Form3 was the onboarding process. It was handled almost entirely by Leanne. No, that's not a quirky name for our People Team... it's our People Lead who had been at the company for approximately a month. To this day, I have no idea how that lassie managed to spin so many plates at once without dropping them all in spectacular fashion. What I do know is that it's a good indicator of how young the company was: our onboarding, background checks, right to work, equipment procurement, travel advice, office access, and so much more were handled by one person who barely had time to settle in herself.

If you have any Agile certifications, look away now - it gets worse.

Every day we had an Engineering stand-up. Yes, an Engineering stand-up with a capital 'e'. The whole business function. At first it was almost a full page on Zoom... then it was a page and a bit. At about this point we quickly realised that what we were doing was nonsensical. Can you imagine fifty engineers giving their daily updates in fifteen minutes? I know some people who would be apoplectically angry about this, and they'd be right - it doesn't work.

Perhaps worst of all from a human perspective was how this translated to on-call. Spoiler: not well. Being on-call meant you were on-call for the entire platform. That's great when the platform is small but the stress grows exponentially with the number of microservices.

Architecting perfectly for scale is impossible - adapting is key.

Growing Up

Within a few months of me joining the company, we were in a whirlwind of change. The teams which already existed - Euro, Platform, UK - were given complete autonomy over their ways of working. The on-call rota was chunked up in a similar way to ensure that: a) engineers were on-call only for their area of expertise; b) engineers could call in expertise if the incident appeared to be caused by an adjacent team's service; and c) the average workload of an on-call shift was stable.

The impact was immediate. We were working faster than ever due to the new focus; quality of life changes for on-call were being prioritised due to the increased frequency of shifts; and teams were adapting ways of working which suited their people, projects and workload.

Autonomy is powerful in the hands of the right people.

One thing of note here is that we did not enforce a framework on the teams. The only standing meeting was our tech town hall for knowledge sharing. At this growth breakpoint most companies go all-in on Agile or SAFe without really thinking it through. You end up with a bunch of Scrum meetings in your calendar which the teams join and tolerate like robots. Don't get me wrong, all of the above are good solutions to the right problem. The important thing is that they aren't a silver bullet for scaling problems. You have to understand which problems you have, which solutions to try, and which solutions simply aren't working.

Fix real problems before presumed ones.

Gateways and Grandmothers

The worst thing about software engineering for me has always been trying to explain what I do. Imagine sitting down to have a coffee with an eighty year old woman, telling her what you're going to be building for the next few years, being met with a completely blank stare, then hearing a confused, "oh... that's nice, son." The things I have worked on in my life simply aren't relatable to the average person: nobody knows what an API is, let alone why it needs to be secured; clouds are fluffy things in the sky; and tracing is a thing you do with a pencil.

I cannot tell you how nice it was to be able to sit her down and say, "when someone transfers money to their friend, it's handled by the thing I made." To see genuine understanding in those wise old eyes. To have her finally believe that I'm making the world a better place. My career has always felt a bit like the maths problems you get in school; when a problem is academic or speculative, it doesn't feel so real.

Having real-world impact is motivating.

People-People

One thing which fascinates me about Form3 is the People Team. As you transition from a scale-up to a well established company, there are so many factors pulling in so many different directions. Most of these, despite being necessary, are undesirable as an employee: compliance, information security, legal, offboarding, post-mortems, regulations. In my experience this is where a company's claim to be "people-centric" falls to pieces. This has not been the case at Form3.

Our People Team are somehow capable of simultaneously doing everything an established business has to do and caring about our people. Based on meeting them I can only assume this is sheer stubbornness and refusal to compromise. Whatever the reason, it's a beautiful thing (just like my RH Mereo 220).

We call organisations "companies" for a reason: it's all about the people.

Onwards

I would love to wrap this up with a paragraph about where Form3 is heading. You know what, though? I really don't know and I love that; I trust that our ethos will keep us right whatever happens. So, in summary:

1. Architecting perfectly for scale is impossible - adapting is key.
2. Autonomy is powerful in the hands of the right people.
3. Fix real problems before presumed ones.
4. Having real-world impact is motivating.
5. We call organisations "companies" for a reason: it's all about the people.