Aparna Krishnan talks open source research, finding structure in uncertainty, and bro culture in blockchain
Aparna Krishnan is a Co-Founder of Mechanism Labs and Thiel Fellow previously conducting research with Professor Elaine Shi working on randomness generation protocols. She was the Head of Education and Executive VP at Blockchain at Berkeley and founded the Education Department at Blockchain at Berkeley. She has taught the world’s largest university accredited blockchain course at UC Berkeley. Aparna cares deeply about enabling convenient privacy which she believes is only achievable through decentralization.
What first got you into the blockchain space? What keeps you active in the space now?
I got into the blockchain space because I was passionate about cryptography and was taking advanced math classes at Berkeley. I did Math Olympiad and Computer Science Olympiad in high school, so coming into college cryptography was already something I cared about deeply. When one of my friends noticed I read crypto papers, he was like “why don’t you check out Bitcoin”. What blew my mind when I first read the Bitcoin White Paper was that it didn’t have any crazy proofs. It was really simple and elegant, yet so revolutionary. From there, I began reading more about the technology on Reddit forums and blog posts. Intellectual curiosity got me into the blockchain space, but now I’m at this position where I care a lot about not only blockchain research, but also making research and blockchain technology into something that can actually be used.
What did your initial delve into blockchain research look like?
When I started researching, a lot of it was working with people from Blockchain at Berkeley– there weren’t resources available online. I didn’t know that distributed systems, a field from which Proof-of-Stake drew on so heavily, even existed. A lot of it was just talking on the Slack channels of companies like Cosmos and Tendermint and reading their White Papers. All of them seemed like the Holy Grail of what the future was going to be. It didn’t feel like I was making progress because I didn’t know what progress meant.
When did you really begin to see progress in your research?
The biggest shift was after I met Zubin and Alexis, my co-founders at Mechanism Labs, about a year-and-a-half ago. We started talking to other researchers in the space and professors at Berkeley about Proof-of-Stake and blockchain. One of our really early mentors, Gireeja Ranade– a professor at Berkeley, suggested we read a couple of academic papers and analyze them. These academic papers changed my perspective a lot; I was able to understand a lot more of what people were talking about in blogs or chats. Later, working with Mahnush Movahedi from Dfinity was super helpful in not just understanding what the process of research looked like, but also learning from an expert in distributed systems. Hearing her take on blockchain and cryptography was very valuable.
How can women entering the blockchain space get started with research?
If you’re interested in distributed systems, cryptography, or economics, blockchain research is comprised of each of these fields. If you’re interested in like cryptography I found Professor Elaine Shi’s work to be really helpful– I’m really glad I had the opportunity to work with her. I would also recommend reading the Mechanism Labs blog for fundamental concepts and gaining an understanding into how you can connect cryptography and distributed systems with blockchain. For people who are more interested in economics, I think Glen Weyl and his group of researchers are doing some really cool work. Definitely follow the Token Engineering community and read up about mechanism design and game theory– all these are really fundamental concepts that will get you in good shape to create different incentive mechanisms. I would watch the Crypto Economics and Security Conference talks or watch Stanford’s BPASE Conference talks. A lot of these researchers talk about different fields and that’s a good way to know who the main people in the field are, and find out what interests you.
How do you find research opportunities?
Definitely reach out to the people doing work that interests you. For example, I met Mahnush at BPASE and I talked to her briefly, but I figured let me cold email her ask her for advice on something and it just turned out that she was so helpful and down to keep working on things from there. You never know when reaching out can lead to something bigger and newer.
Why is it important that women are a part of the blockchain space?
In the blockchain space, especially because we’re so early, it’s important to have a good representation of both men and women. It’s only what we start now that will carry on into the future. When I initially got in into blockchain, I was the only girl in the Executive Board of Blockchain at Berkeley, and oftentimes even though there isn’t explicit discrimination there is an implicit bias. There are all these social events which “bros” can be part of, but you–as a girl– just definitionally don’t fit in. And a lot of discussion and transmission of information happens at these events. Initially it didn’t seem like this was a big deal, but over time I realized that it was hard for me to start a conversation with people at a networking event if I went with a group of guys, because guys just automatically have that “bro to bro” bond. Over time, the women who are in the blockchain in space will start to leave if you don’t bring in other women in. As this space evolves, you want to maintain this balance and thought diversity and the only way to do that– and prevent the people already in the space from leaving– is by being more inclusive.
Mechanism Labs is the world’s first open source research lab. What does it mean for a lab to be open source?
An open source lab just means that anyone who wants to contribute can contribute. The way that research is traditionally done is that only once the paper is publishable it is released, but oftentimes during the stages of the paper researchers could go in different directions or have different ideas. Open source means that every stage of this process is on GitHub or on a common platform where other people can reshape the way that research happens. With Mechanism Labs the audience is people who are interested in academic research and the goal could be publishing in a journal, creating a paper that other people can reference in the future, or speaking at a conference depending on the specific paper.
Why is this model for research powerful?
We created this open source model because we noticed that when we were doing research at Berkeley we could only work with other people in Berkeley and at a company you can only work with other people in that company– i.e. these institutions created barriers of idea sharing that we hope to break down.
The lack of structure in your schedule and uncertainty that stems from becoming a Thiel Fellow seems pretty different from going to Berkeley with a strict class and assignment schedule. How has the transition been?
That’s a great question, and I’m about to release a blog post that covers that. To give you a high-level answer, I definitely prefer having structure that I create for myself. Even though a lot of my time is unstructured, I create enough deliverables and forcing functions to ensure I’m making progress. For example, if I want to get a coding sprint done then I’ll tell a few friends and then I just feel accountable. Having Zubin and Alexis helps to create structure– for example, every week we have meetings at designated times and fixed locations.
At Berkeley I had no time to think. It was a year where I filled up all my waking moments with work and I felt bad sleeping or taking a break because there was so much work to do. Going from that to taking the Thiel Fellowship, where I didn’t have deadlines was a really nice break. But I kept running into questions like “What kind of work I should be doing?” and “What’s valuable work, and what’s not?”. I realized that I had to generate small amounts of output in a specific direction to determine if the work I was doing was valuable– for example writing a blog post on a topic and getting feedback from the community was a much better indicator of value as opposed to just thinking about if exploring the topic was valuable in my head. Maybe you don’t need to write a paper about a topic, but writing a whole tweet thread will give you enough feedback to see if that’s a good direction to pursue, or if you need to make adjustments.
I think everyone might have to decide how to deal with a lack of structure at some point in their life, but it’s a different point at which different people face this problem. It’s a nice thing to be able to decide what to spend your time on.
Connect with Aparna on Twitter @aparnalocked and read her work on Medium.
See you next week when we chat with Linda Xie! Write to Roshni Rawal at roshnirawal@berkeley.edu.The she256: Fireside Chats are sponsored by Upscribe.
