The 2nd Annual Modern Data Stack Conference is back on September 22 - 23, 2021. Learn more and register here.

HomeBlog
Three Months Under the Bodhi Tree: The Fivetran Research Rotation
Fivetran
HomeBlog
Three Months Under the Bodhi Tree: The Fivetran Research Rotation

Three Months Under the Bodhi Tree: The Fivetran Research Rotation

A unique program allows any Fivetran engineer, anywhere in the world, to spend months pondering one interesting technical problem.

By Meel Velliste, August 18, 2021

At the beginning of every financial quarter, a few Fivetran engineers get to do something unusual: take a three-month break from their regular work to think deeply about one interesting technical problem, in close collaboration with me, the CTO. Success isn't guaranteed — my office often explores unconventional ideas — but any solution we come up with is likely to have a profound impact on our product and business.

This is the Fivetran Research Rotation, and any Fivetran engineer anywhere in the world can apply for it. One recent participant said that it had been “my most rewarding time at Fivetran, by far.” Another likened it to “meditation under the bodhi tree," and said that he "did not know how enlightened I was until I got back to my home team."

As Fivetran CTO, I love the program because it enhances the quality of our experimental research while giving our engineers an incredible creative and intellectual outlet. In this post, we look at how the program originated, how it works, and why it’s essential to our technical progress and engineering culture.

Deeply Collaborative Research, Continual Innovation

At Fivetran, we never want to stop innovating and pushing the technological envelope. How do we make that happen, though? It’s not enough to simply respond to customer feedback and iterate gradually, as crucial as that is. To truly push the limits, we have to invest in ambitious research projects that force us to take big technical leaps. If we don’t do that as a company, we risk growing stagnant, technologically. We need to see innovation as revolutionary, not just evolutionary:

My team is focused on this kind of revolutionary innovation, and we could be a closed shop — a kind of elite unit of intrepid engineers working at the cutting edge. But Fivetran employs hundreds of engineers with direct experience of the problem space, and each of them has a unique perspective and skill set. Why wouldn’t we want to tap into their expertise and creativity?

The Fivetran Research Rotation allows us to select a few engineers at the beginning of each quarter and add them to the team. That gives us a constant influx of talented new people with original ideas and different ways of looking at problems. It makes the research environment more dynamic and intellectually diverse, increasing the chances of real technical breakthroughs.

It works the other way, too — the engineers go back to their home teams with the perspective that comes from sustained research and a more abstract, comprehensive and strategic way of thinking about the product. Ultimately, the Rotation drives a rich cross-pollination that makes us all better engineers and researchers.

A Transformative Experience for Engineers

Time is probably the greatest advantage of the Rotation. Instead of working on discrete issues across two-week sprints, engineers can focus on one problem for several months. We might not see the results of our work for a year or more; individual projects can easily span multiple rotations. This allows for many iterations on a single problem. 

Rotation engineers keep a journal to track the evolution of their thinking, and we have a daily working session so my team and I can brainstorm with them and offer guidance. With plenty of time and intellectual support, one engineer can make an incredible amount of progress in just a few months. Rotation engineers often say they feel like they’re moving “at light speed.”

I want the engineers in the Rotation to become more creative, more curious and more collaborative — to adopt the mindset that every problem has a solution, no matter how hard, and what seems impossible to one person is often quite possible for a team. That’s one of the main reasons why, at Fivetran, we never want to accept inferior solutions.

Engineers tend to come out of the Rotation with a fresh perspective on our technology and software engineering. Here’s what two recent participants, Levon Korganyan and Abhijeeth Padarthi, had to say about it.

Fivetran Engineers on the Value of the Research Rotation

Levon Korganyan, Staff Software Engineer, Oakland

This was, personally, my most rewarding time at Fivetran, and I’ve been here three and a half years. I’ve been evangelizing this to all my engineer friends, both at Fivetran and outside it — because part of the worry, for every engineering candidate, is, “Am I going to be allowed to be creative? Will I be able to put my own spin on things?” Code is very standardized, coding teams run in sprints, the product might be fully scoped — you take the ticket, you work on it. Fivetran generally has a good balance of structure and creative freedom, but the Research Rotation takes it to another level entirely. Engineers in the Rotation directly contribute to the product vision. You have real creative control over your research, but you also have incredible guidance every day. You explore the literature, and you learn to step back and think innovatively about the problem you’re facing. This is a skill you can take back to your home team — you don’t have to be a researcher full time to step back and think creatively.

Abhijeeth Padarthi, Software Engineer II, Bengaluru

Honestly, the Research Rotation was like this period of meditation under the bodhi tree, and I did not know how enlightened I was until after I got back to my home team. I say this because at first the problem I was working on seemed familiar and simple, but working on it every day pushed my technical knowledge to the limit. We had to deal with every single data connector being written at Fivetran — not just one technical detail of one connector. And the CTO team pushes you to look at every problem from different perspectives. What are all the relevant commonalities and all the key differences? Most developers don’t deal with the problem this broadly, and it became this very abstract and much larger problem. It gave me a deeper understanding of what we’re actually trying to do at Fivetran. Now I understand how all the parts of our system relate to one another, and that has changed how I approach problems as an engineer.

Join Our Engineering Team!

If you’re interested in the Fivetran Research Rotation — and an engineering career that offers both technical challenges and great work-life balance — please consider joining our team. We’re hiring across multiple engineering positions, and my office is currently looking for a research and development engineer, a position that can be based in the Bay Area or fully remote.

Start analyzing your data in minutes, not months

Launch any Fivetran connector instantly.