Unleashing Data Compatibility with SQL: An Interview with Brennan of Kwil
Kwil is the first permissionless SQL database for the decentralized internet. Their mission is to provide the infrastructure to power data storage and retrieval for the next generation of Web3-enabled applications.
Can you tell us a little about your background and what led you to start Kwil?
I have always been pretty entrepreneurial. When I was 13, I made money by arbitraging items in a popular video game and managing payments through PayPal. What began as a couple of bucks on the side turned into a decent amount of money and ended with me getting banned from PayPal for violating their ToS by lying about being 18 (they let me keep the money, but to this day, I am still banned).
When I was 16, I started a company that did landscaping and excavation work in the greater Boise area. After briefly mowing lawns and pulling weeds, I evolved the business to handle more complex (and higher profit) tasks like rock beds, irrigation lines, and demolition. Over the next three years, I grew this company to 14 employees.
In 2020, this business stopped after a white-water kayaking accident resulted in me being medevaced to the ER. While on my hiatus from manual labor, I spent my time pursuing a passion I had recently picked up in my first year of college: programming. I spent a lot of time over the next eight months creating projects I thought were cool: financial derivatives analysis, web scraping, and even natural language processing applications.
In Fall 2020, I took a gap year from my (then online) university and took an internship with a general contractor. Throughout this internship, I spent countless hours continuing my passion for programming, many weeks working the minimum hours at my job so that I could write code instead. During a particularly slow day on the job in December 2020, I went to Barnes and Noble on my lunch break and bought a book on Solidity. A month later, I found my way into the Arweave ecosystem while trying to build a decentralized social protocol and got invited to present what I was working on at their first ArConf. It was there that Rob Chan from 4SV would come to me looking to write an angel check, resulting in me leaving my internship early to focus on building Ecclesia/Kwil.
Talk about the original product visions for “Ecclesia” and “Kwil Social”? Why did you ultimately decide to focus on building infrastructure instead?
Ecclesia originally started as something I was building for fun. Once I received some funding and knew I had to make it a business, I began looking for a problem it could solve. Initially, I set my sights on censorship resistance. I began with basic blockchain immutability but eventually tried to implement HTTP tunneling technology to allow access in countries that faced government censorship.
Unfortunately, it’s extraordinarily challenging to build a business on that premise. Next, I found myself trying to make the social application a DAO communication platform (pitched as an alternative to Discord). At this point, we made the decision to rebrand to Kwil, and adjusted our branding to be more playful, aiming to appeal to the DAO audiences we targeted. It was really a solution in search of a problem, and frankly I did not understand the needs of the customers I was targeting.
By October 2021, Kwil Social struggled to get meaningful traction with some of the DAOs that had agreed to be our first cohort. Users occasionally came on the platform and made a post, but generally they continued using Twitter, Telegram, and Discord for their communication. I was desperate for any meaningful traction to validate that we were building something valuable.
While on a call with a DAO building an NFT art gallery, I could tell they were not interested in using our product, so I began shifting the conversation to see if there were other needs I could pinpoint and build features to solve. As one of their engineers was covering their architecture, I asked why had decided to build on AWS instead of building on a decentralized platform. They informed me that they had tried Arweave and Filecoin but could not store and query their data in the ways they needed. Desperate for some meaningful use, I brought up how we had built a basic database layer on top of Arweave and asked if they would be interested in using it. They agreed to try it and became the first users of Kwil the database.
For the next six weeks, we continued with social being our main product. After the six weeks, we reflected on how we were still largely unable to convince even smaller DAOs to use the platform and yet had many other projects who came to us asking to try the database. We officially pivoted shortly after that, and it has undoubtedly been the best decision we have ever made.
Can you talk through the data storage and indexing issues in Web3 today?
Currently, it is nearly impossible to build highly performant, truly permissionless applications that contain complex user functionality. Many Web2 technology products we consume are powered by robust data stores that enable intricate data to be managed seamlessly. In Web3, this simply does not exist. While blockchains like Ethereum and Solana enable key-value stores, and Arweave and Filecoin act as high-capacity file systems, there is no solution for managing complex data structures in a permissionless environment.
Generally, Kwil doesn’t do much for blockchain data indexing (however many of our partners do). The real breakthrough with Kwil is the ability to write data that coalesces around a single schema permissionless-ly. With smart contracts, the world witnessed the power of permissionless data writes and composability. Whether through a smart contract that enables token lending as a single application or an ERC20 contract that seamlessly incorporates into tons of applications, DeFi thoroughly proved the unique value of permissionless data writes. Because Web3 is expanding to more data-intensive sectors like social, science, and gaming, Web3 infrastructure also needs to support the data needs of these new verticals. The role of Kwil is to support the complex data structures required by these new applications.
Describe the vision for the Kwil protocol and how it simplifies the decentralized application development process.
Although Kwil is built for complex data sets, we believe that the process of managing such datasets does not need to be complex. Kwil is building something that doesn’t have existing counterparts in Web3. In building this unique offering, our goal is to make it as easy as possible to build on Kwil. A great model for how this has been done previously is Solidity.
When you first learn Solidity, it can be slightly confusing. Solidity has new concepts, rules, and best practices that are not common in other languages. However, once these concepts are understood, developing smart contracts is easier in Solidity than in almost any other language. Anyone who has written a complex smart contract in Solidity and then tried to write an equivalent in JavaScript knows that it is unexpectedly hard to produce the same result.
Solidity is so much easier to use because it is not just a language, but a tool for building smart contracts, which are an entirely new type of application. It contains features like contract inheritance, function modifiers, and require/assert statements that make Solidity uniquely effective at building permissionless applications and fulfilling the value of Ethereum. It is certainly possible to build the functional equivalents in other languages, but it is more complicated than in Solidity. In this same way, while Kwil does require some initial overhead in learning how to utilize it properly, it ultimately makes it easier for developers to create new types of applications enabled by it.
Why the decision to focus on SQL (Structured Query Language)? Why does SQL make the Web3 development process feel more native to Web2 developers?
SQL databases make up 60–73% of production databases in Web2. For a developer coming from Web2 and looking to take part in building the next version of the internet, it’s important to have familiarity when they are building their applications. While DDL on Kwil has its own unique language (we are aiming on releasing it this year), querying data from Kwil can be done with traditional SQL, making the application development process much more efficient and intuitive.
Why is the mass adoption of decentralized technology partially dependent on the ability to provide frictionless developer experiences?
The single most important aspect necessary to drive Web3 adoption is building phenomenal user experiences. Products need to be quick, intuitive, and secure. Currently, too many Web3 products suffer from poor consumer experiences. Whether it be something complex like the risk of wallet hacks and loss of consumer funds or simply requiring a user to keep track of a mnemonic and use a browser extension, Web3 has a serious UX problem.
The only way to solve this issue is with great engineers. While designers, product managers, and marketers all have a role to play, it’s ultimately the job of engineers to build solutions to fix the poor UX that is endemic in Web3. In order to encourage these developers to make the jump to build in Web3 and to ensure that their time is well spent, infrastructure companies must provide great developer experiences. Companies should not just focus on shipping innovative technology but need to emphasize building full products. Documentation, extensibility, and technical support play a huge role in encouraging the adoption of a product.
How does data composability fit into this picture, and what does composability ultimately unlock?
One of the key breakthroughs in Web3 technologies is composability. Whether through smart contracts on Ethereum, Permaweb applications on Arweave, or databases on Kwil, the ability for developers to use each others’ work as building blocks is an advantage that cannot exist in Web2. Being able to permissionless-ly plug into both applications and the data moats surrounding them using only a few lines of code is an extremely powerful force for encouraging technological collaboration, as it disincentivizes destructive competition and makes constructive collaboration trivially easy.
A great way to think about data composability is to view it as the next iteration of open-source technology. The biggest issue with open source is that it lacks foundational incentives for maintenance and upkeep and encourages free-riders to utilize software while contributing nothing back. While open source allows developers to plug into existing code, data composability also allows them to plug into the data already tied to that code, while still mandating that they pay for protocol usage like anyone else. Since so much of software’s value is held in an application’s data, permissionless data composability means that external users are more intimately aligned with the incentives of the underlying software and community.
What are some of the most compelling use cases for Kwil?
One of the most compelling use cases for Kwil is in decentralized social applications. There are few industries with such asymmetric value capture as social media/content creation, and I believe Web3 is the perfect technology to solve this. Currently, Kwil is working with social projects covering various subject areas, including social graphs, messaging applications, and interest graphs. Many of these applications prioritize returning value and rights to the consumer, effectively removing the incentives for abuse that exists within the social sector today.
Another sector that I am personally excited about is decentralized science. Many inefficiencies currently exist in research due to unnecessary middlemen and the incentives surrounding research funding. Some really cool decentralized science projects are aiming to remove these antiquated middlemen and provide alternative incentives for research and study replication. While this is an area that I am really excited to dig a bit more into, there are already a couple really awesome projects in stealth that we are working with who have a deep understanding of these issues and some truly innovative solutions for them.
Can you talk about the rest of the team and who is building this alongside you?
The Kwil team is an awesome mix of young, fresh-minded builders and senior, deeply-knowledgeable industry veterans. Our team is in-person in Austin, Texas, which has allowed us to move quicker and develop a strong team culture.
My older brother, Luke, has been working alongside me and brings many of the skills that balance my own. While I tend to be a bit more on the rebellious side, Luke is the type to color-code his sock drawer. This dichotomy of personalities has been critical to our success thus far, with Luke dealing with anything related to operations, finances, legal, HR, or accounting.
Also on the team is Luis, a college classmate who I enlisted to help me expand our user base. Growing up around serial entrepreneurs, Luis has a passion for building innovative products and solving hard problems. Mixing this with his prior experience in business development for database companies, Luis has become a critical part of day-to-day operations, handling everything from Kwil branding and external image, to user testing and G2M roadmap and execution.
On the more seasoned side of our team is our CTO, Randal. Randal has an extensive background in early-stage startups, cloud scale development experience in streaming, storage, messaging, and SaaS platform development more generally. After founding two startups that were both acquired, Randal spent 14 years as a VP of Development at Oracle. Randal is responsible for building critical pieces of infrastructure, especially regarding event-sourcing, concurrency, and system resiliency.
Finally, the fifth and most recent member to join our team is Bryan. Bryan has deep experience in building resilient microservices, and has been a leading engineer for both startups as well as large companies. Bryan is primarily responsible for the design and implementation of our cloud architecture, as well as building the Kwil DDL language.
Congrats on closing the $8.9M fundraise! How will the proceeds from the round be used to bring the Kwil vision to life?
Kwil is amply prepared to build aggressively over the coming years. While many companies are taking their feet off the gas and preparing for a potential recession, Kwil has the firepower to continue growing over the next several years; however, we can also be flexible with runway to allow us to withstand any sort of economic downturn.
The funds will go towards scaling the team as our customer market demands. If we find that growth is strong and we can quickly build a sustainable business model, we can grow the team aggressively. On the other hand if tech investment continues to cool off, and broader adoption of new technologies slows, then Kwil can keep our team nimble and resilient and grow at a more moderate pace that is in step with the market.
What does the Kwil team need to get right in the next 3–6 months?
I am extremely excited to release our next version of Kwil. Now that we have initial feedback from our first live version, we have built some phenomenal functionality that is truly unique to the industry. When these features launch in our next version, it is critical that we continue to listen to our early cohort of users to ensure that the features that we are building are needed and applicable to a broad market. The state of the Kwil product is still highly malleable and could continue to evolve in various ways. The next six months will be pivotal in solidifying our product direction as we respond to the needs of our users and look to expand the functionality available in Web3.
How can folks learn more about Kwil and use the protocol themselves?
The best way to learn more about Kwil would be to follow us on Twitter (https://twitter.com/KwilTeam) and join our Discord (https://discord.com/invite/n7SKUMQnAs). These are the first places we release news and information about the product.
As for how users can use Kwil, builders can find examples of how to use the MVP on our Youtube (https://www.youtube.com/channel/UC7byhahd6AZ0zPSjFX5NTlQ). Builders can also easily view databases with our DB visualizer at db.kwil.com. Since this is an MVP, we encourage builders to stay around for our next version, which will be launching in the near future. We are actively looking for beta testers, partners, and new team members to join us in building this foundational technology. The easiest way for any contact is to email info@kwil.com, or you can find me directly at brennan@kwil.com.
Our team could not be more excited about the direction we are going, and we look forward to building the future of the open internet with the rest of the Web3 ecosystem.
