Sophisticated Canaries: Singlepane Tells “What Happened Was” in Grafana

I was going to write about how “Being Right Doesn’t Matter” because I believe that recognizing that is a turning point in an engineer’s career. Let go. The most impactful lessons for a team are learned through trial and error. Extending trust to your teammates is what builds bench strength – yada yada. Actually, I did write about it on the night of January, Friday the 13th and finished the full post. It’s sitting in draft because I hadn’t found the best header picture and spent time in vim to edit wp-config.php and a bunch of other stuff with S3 buckets so…what happened was…

Alan Page wrote it first. Read his: he’s got 20 years on me and it’s better anyway. We could never work at the same company and still land on the same virtual page. More importantly, now I have to 1-up that with the other post I had unfinished because I’m trying to get better at timing.

Meet Singlepane, a literal garage startup that is building “What happened was…” in Grafana. They create dashboards for large-scale world events using public endpoints, open-source intelligence, and craft sophisticated canaries that tell the best stories on the backbone of the internet. They reverse engineer time. They use the past, present, and future across and over milliseconds. They don’t post-mortem an outage: They post-mortem the context. For example, this is what happened in Kharkov, Ukraine, in February when Russia attacked and is the type of work that only strong founders could truly produce as it requires you to be vigilant, hyper-aware, timing-centric, and fearless. Singlepane shared this as realtime OSINT (open-source intelligence).

Singlepane wanted to explain the cyber attacks in Kharkov against troop movements for the OSINT community. Sourced from Singlepane. Used with Permission.

Currently, Singlepane is known in the devops community as “The dudes that built Deadbird” – a set of dashboards checking all the public endpoints for Twitter’s distributed microservices stack after a large part of Twitter’s team was let go. There was a lot of fluff and people saying Twitter was going to die – that’s pretty binary if you know infrastructure. We can assume it was built well. I know too many who have this bookmarked to learn from this – it’s an educational moment. Why build on microservices? Well, simply, it would take a lot for them to actually die when architected as highly available. The question is: what is a lot if we’re talking about…decisions and a series of choices? That’s fascinating. Let’s see what happens.

Deadbird only represents something Singlepane has built in the last few months, but to understand how this team got here and where it will be, one has to rewind 12 months to when I was privileged enough to met them. Meeting a team of founders when they are still nameless and knowing “You’re about to make something because you are inspired by the contexts you see to help.” Almost a year ago I discovered two founders through Twitter by following their work on the Ukraine war. I realized after a few months they had also worked in games infrastructure, IPTV, and authentication. They were technically skilled and definitely understood how to look at timestreams.

There was no company. They didn’t have a name. 5 people followed a tiny Twitter account at the time – I was one of them. I had to know. Who was behind these graphs? But they were so discrete that I struggled to connect the who with the what and finally DMed them. That’s not normal for me (and I’m no longer on Twitter either – only a shell that I monitor for activity).

Battle of Kharkov – Local Internet Infrastructure. Sourced from Singlepane. Used with Permission.

I waited to ping or say anything too much Feb – March ’22 – I thought for a while Singlepane was trying to stay anonymous for obvious reasons. Then, Singlepane responded to my Twitter DMs.

I started sharing their amazing posts – Singlepane was showing why and when specific cybersecurity incidents happened in Ukraine. I messaged them something along the lines of “I need to know who you are. The way you connect world events with infrastructure and decisions. Not everyone thinks like that.” A few months later we hopped on Discord after I told them where I’d been. Sad me said “I’ve done the startup thing before and I’m tired. I ran a startup for 6 years – I want to stay in games enterprise I think and still figuring it out. But I think I can help you get started.” A few conversations and queue shipping some cash and introducing them to my attorney in July ’22.

The co-founders of Singlepane, currently in stealth and have asked me to remove their names, have deep experience at the intersection of site reliability and auth, aren’t only building Down Detector for adults – they’re working on their Product-Market fit for their stories and it can go in several different directions. They are great builders but building slow – right now isn’t the best time to do a raise, I told them as much last year, and when you’re talented you can afford to wait especially if what you are building is flexible and you are practicing active listening. They are adaptable and what is truly priceless is their brain and how these founder stories. They are two best friends who enjoy being around each other but whose paths have taken different turns and who want to continue to contribute to the engineering ecosystem transparently.

Wider view of Kharkiv internet. Sourced from Singlepane. Used with Permission.

Singlepane, I look forward to your success because I know, like you do, that eventually your own timestreams connect with exactly what you need and as long as the universe doesn’t wink too weirdly, you will earn it. Keep building cool dashboards that unite all the timelines, not only the darkest ones. So much of infrastructure is looking around corners and we need more fearless leaders like you both.

Singlepane currently has two great stories for which they’ve demonstrated their incredible capabilities to use publicly available information to explain what happened. The topics they research are most valuable to press and the public – But they need another good story. The third time is always the charm. The question is when and exactly who will it help.

They are still looking for that third co-founder, but have plenty of time to evaluate who that should be as they experiment with their stories. As for me, I will continue to cheer them on regardless of where their careers take them and mostly feel lucky I have had a few opportunities to sit in the room as an active listener who truly believes they are some of the most interesting people I know.

And maybe if I got it right this time, I nailed the timing for the person who needed to read this to help them and I don’t yet know who you are. As for me? I love my team so I won’t be going anywhere anytime soon.