Fuzzbuzz, a graduate of the most recent Y Combinator class, got the kind of news every early-stage startup wants to hear when it landed a $2.7 million seed round to help deliver a special class of automated software testing known as fuzzing in the form of a cloud service.
Fuel Capital led the round. Homebrew and Susa Ventures also participated, along with various angel investors, including Docker co-founder Solomon Hykes, Mesosphere co-founder Florian Leibert and Looker co-founder Ben Porterfield.
What Fuzzbuzz does specifically is automate fuzzing at scale, says co-founder and CEO Andrei Serban. “It’s a type of automated software testing that can perform thousands of tests per second,” he explained. Fuzzbuzz is also taking advantage of artificial intelligence and machine learning underpinnings to use feedback from the results to generate new tests automatically, so that it should get smarter as it goes along.
The goal is to cover as much of the code as possible, much faster and more efficiently than human testers ever could, and find vulnerabilities and bugs. It’s the kind of testing every company generating code would obviously want to do, but the problem is that up until now the process has been expensive and required highly specialized security engineers to undertake. Companies like Google and Facebook are able to hire these kinds of people to build fuzzing solutions, but for the most part, it’s been out of reach for your average company.
Serban says his co-founder, Everest Munro-Zeisberger, worked on the Google Chrome fuzzing team, which has surfaced more than 15,000 bugs using this technique. He wanted to put this type of testing in reach of anyone.
“Today, anyone can start fuzzing on Fuzzbuzz in less than 20 minutes. We hook directly into GitHub and your CI/CD pipeline, categorize and de-duplicate each bug found, and then notify you through tools like Slack and Jira. Using the Fuzzbuzz CLI, developers can then test and fix the bug locally before pushing their code back up to GitHub,” the company wrote in a blog post announcing the funding.
It’s still early days, and the startup is working with some initial customers. The funding should help the three founders, Serban, Munro-Zeisberger and Sabera Hussain, to hire more engineers and bring a more complete solution to market. It’s an ambitious undertaking, but if it succeeds in creating a fuzzing service, it could mean delivering code with fewer bugs, and that would be good for everyone.