Business, tech, and life by a nerd. New every Tuesday: Splitting Light: The Prism of Growth and Discovery.
Share
Splitting Light: Season 3 - Episode 02
Published about 2 months ago • 4 min read
Splitting light
Season 3 Episode 02
What now @ Scaleway
If you are no longer interested in the newsletter, please unsubscribe
In June 2019 I understood that Scaleway was my best option to work on something truly interesting. During the preceding weeks, I had taken my work load more lightly and I was more rested. I finally messaged HR that I wanted to talk with Arnaud (a).
The meeting was scheduled and we stayed, both of us, in a meeting room for several hours. First reminiscing. We talked about the hardware lab, the Carbon14 product. Then I talked about the storage team, the products, what I liked and what I didn’t like. We sought to find a product or line of products that I could work on, something that would benefit both Scaleway and me.
Me VS one of the ceiling columns in the underground cafeteria of the office (1)
We round up three options. Three places where there would be significant challenges.
The first was to create Scaleway’s lambda equivalent product. There were many technical constraints, some were very hard. The product was at the very top of the cloud stack. Everything underneath had to work flawlessly. There were hard performance constraints. Arnaud mentioned some prior work done by Folays (b) which could be used as a supporting base. It looked very promising.
Another option was to join the Kubernetes team at Scaleway to work on the Kubernetes product. It was a very hype product and there were a lot of opportunities on that line of product. Multi-cloud, multi-clusters, failover, and many more. It was the sheer amount of work that was attractive.
The third option was to join the registry product team. Registry is a product where you store container images, similar to storage but very tailored to a specific use-case. The challenge there was to optimize for latency and availability. Multiple products depended on the registry. Kubernetes being one, but also the lambda product in some sense.
We concluded the meeting. My homework was to think about the possibilities and do some research on my own. I quickly called Folays and we talked about the work he had done. I did quite a lot of personal research. Looking at the possibilities, the technical challenges, and what had to be done. My general impression and intuition about the options. I eventually came to the conclusion that, in addition to liking the storage products, I really liked the storage team. I felt very close to the team as a whole.
I was attached to my team mates. I was attached to the products we had built. Object storage was in a sense my baby. Carbon14 was the last physical artifact of the lab. I decided to stay in storage. I decided to continue the path of storage. But this time, I decided to try a different path, I wanted to have the product owner role. I messaged Arnaud my conclusion and shifted into the new role.
Carbon14 boards in the DC4 bunker before assembly (1)
I had high hopes! I wanted to lead the team better. A few people had left. Seven of us had stayed. Me, Folays, Florian (c), Louis (d), Quentin (e), Albert (f) and Nicolas (g). The rest, Théo (i) , Alexandre (j), Maxime (k), Florent (l), Loic (m), Marian (n) had left for other adventures.
Carbon14 chassis husks in the DC4 bunker before assembly (1)
In a strange mind twist, I didn’t think that we had already accomplished great things. The new role was, in my mind, specifically to push forward, to light up a path to success. In the last year and half, we had not launched two products, multiple regions, maintained multiple legacy products? No. This was not an accomplishment. How could I think that? Maybe it was because we didn’t celebrate the success. Maybe I was trying to fill a void that I could not fill with tech? At the time I could not fathom my growing footprint in Scaleway. I never did enough to satisfy myself.
My own idea to “accomplish more” was more structure for the team and product. I started with documenting more and organizing the tasks better. This is where I stepped in first.
(1) Photos by me
(a) Arnaud de Bermingham, Founder and CEO of Scaleway at the time, now CEO of OpCore (datacenter focus spin-off of Scaleway)
Splitting light Season 3 Episode 08 Recruiting more floppys If you are no longer interested in the newsletter, please unsubscribe Now that the team had settled back into an equilibrium, we needed to replenish our ranks. The storage team, after the departure of several people, was down to seven people. Split half and half into senior engineers and junior engineers. We needed to recruit more just to bear the workload. We had seven legacy storage products, two new in GA and more in R&D....
Splitting light Season 3 Episode 07 A second world first If you are no longer interested in the newsletter, please unsubscribe July 2019 to February 2020 In February 2020, we, the storage team at Scaleway, were released a world first. Personally, it was my second, the first one being the bare metal arm cloud back in 2014. Image from the blog post on the feature (1) We released “the first non-AWS S3 compatible GLACIER storage class”. This was a year after the first region in General...
Splitting light Season 3 Episode 06 A learning from Gaspard: delegate If you are no longer interested in the newsletter, please unsubscribe End of July 2019 In August 2019 Gaspard (a) joined Scaleway to manage the storage team. This was a big change for me. Gaspard was the last person at Scaleway that I consider a mentor. DC4 building decorated with banners for Scaleday. (1) Although he had prior storage experience, his experience was on consumer facing devices. He had worked for Lacie which...
