Why Rockset Is My Next Job After Facebook
December 1, 2021
“At every step in this process, I’ve been consistently impressed by the quality and caliber of the team."
On Monday, I joined Rockset.
I am joining Rockset as its first director of engineering and first external manager hire. I come here from Facebook, where I spent the last 10 years building and supporting teams. Most of my work was in the core C++ libraries and distributed systems components that power Facebook’s infrastructure. The teams worked on abuse detection infrastructure (like Haxl), RPC and service discovery systems (like Thrift), and a wide range of the core libraries like jemalloc and Folly. I helped bring address sanitizer to Facebook, fixed a lot of bugs, and wrote Haskell in production.
But now I’m at Rockset. Let me tell you why.
The best reason to join any endeavor is always the people. And this is a group of people I really, really want to work with. Joining Rockset is a reunion of sorts for me. One of the benefits I gained from spending so much time working on core libraries and tooling was getting to wander around a large, complex codebase and work with a huge range of people solving their own very difficult problems. Venkat’s team built TAO, the optimized graph store that powers FB. Dhruba and his team were working on
RocksDB. Tudor was one of the most prolific engineers in the codebase, having built half of the
Folly library that powers almost everything at Facebook and Rockset. Nathan wrote the omnipresent
F14 hash tables, and the multi-producer, multi-consumer queue that’s at the very heart of Thrift’s handoff between IO and CPU threads.
When it comes to Rockset, though, it’s not just the old friends that made me want to come and work here. At every step in this process, I’ve been consistently impressed by the quality and caliber of the team. It’s difficult for me to overstate how strong this engineering team is and it’s very easy to join something that you feel that way about.
The second reason I’m really excited to join Rockset is that the company seems to be at a critical and very interesting moment in its maturation. It’s all well and good to solve some very hard database problem, or craft the finest of C++ templates, but ultimately at a company you need to build things that are useful to people, and usable by people. In my estimation, Rockset is entering a defining era where it will be imperative to make this novel technology both useful and usable to ever-growing numbers of users.
Everything feels like it’s accelerating. This creates the most interesting challenges, technical and otherwise. It feels like a ride I want to be on.
The final reason is a bit more personal. The mix of work I think I’m likely to encounter in this role is a great mix of things I understand and am comfortable with, and a mix of things where I get to be a proper newbie and ask all the dumb questions. I was looking for a space where I get to do some of the things I like to think I’m good at, as well as challenge myself in new and interesting ways. A little, and just a little, discomfort is good for the soul.
I am super excited to be here and, with all that said, super excited to get to work.
Real-Time Analytics At Lightning Speed