I joined StarLeaf in 2010 when there were only a handful of employees. I had no worries about joining such a small company, as I had worked with some of the team at previous start-ups and I knew what we were capable of.
Back then, we were already planning to create a video calling and conferencing system with a range of our own hardware endpoints. However, the StarLeaf platform was no more than a twinkle in our founder's eye. In those days, we planned to design, manufacture and sell a Video Private Branch Exchange (PBX). This was embodied as a range of on-customer-premise video infrastructure devices. To this day, references to the PBX are still apparent in our code base.
I was hired as a Hardware Engineer, based in our Rickmansworth office, to help develop the PBX. There was a plan to neatly divide the development effort, with endpoints in Shelford and infrastructure in Rickmansworth. The reality turned out to be more blurred and I worked on hardware for endpoints as well as infrastructure. This yielded diverse projects: endpoints are mostly about lots of human-friendly interfaces, without blowing the cost budget, whereas infrastructure is mostly about lots of raw performance without blowing the thermal budget. From my point-of-view, this was great. Variety keeps the job challenging and fun.
A few years ago, it became clear that customers were reluctant to invest in infrastructure and the StarLeaf platform was conceived. Furthermore, off-the-shelf hardware (full of general-purpose Intel CPUs) had become the most cost-effective way to create video infrastructure. Our own infrastructure hardware projects were therefore cancelled. With the original plan in mind of endpoint development in Shelford and infrastructure development in Rickmansworth, it was suggested that I might like to try my hand at software.
Whilst I had dabbled in software before, I had not done any serious projects and was very much a novice (I now know that the technical term is ‘noob’). Fortunately, it turned out that we have some very patient people in our software team and they soon got me creating useful code. Having reached a bit of a plateau in hardware, it was refreshing to be back on a steep learning curve.
In recent years, the number of hardware projects has been increasing again and it became apparent that we needed a little more hardware engineering bandwidth. One of the smaller hardware projects came my way, named Tyne. This evolved into the product that we now know as Pronto. Over the last year or two, thanks to Tyne and a few other hardware tasks, I have been able to share my time between hardware and software. This is the ideal mix for me. As I said before, I feel that the variety keeps the job challenging and fun. I suspect that there are not many companies where I could get the diversity of roles that I get in StarLeaf. Nor would many companies (or their staff) be happy to train me up from almost zero in something as complex as our software.
Long may the fun continue.