Using AI to help developers is not an original idea. Nowadays you have startups like DeepCode offering AI-powered code reviews and tech giants like Microsoft working on applying AI to the entire application developer cycle. Still, Kite is not a newbie.
Kite debuted its offering privately in April 2016 before launching its developer sidekick powered by the cloud publicly in March 2017. The company raised $17 million in January 2019 and ditched the cloud to run its free AI-powered developer tool locally. Now it is dipping its toes into monetization with a paid version. Kite comes from Adam Smith, who founded Xobni, an email service launched in September 2007 that Yahoo acquired in July 2013.
“These are just the next steps on our journey to revolutionize how developers write code using AI,” Smith wrote today. “We are just scratching the surface of what machine learning can do for developers. We have a long way to go, and to make sure we have the revenue to get there we’ll be building out a portfolio of paid products targeting professional developers.”
Python: Kite Free and Kite Pro
Kite Pro is the company’s first paid product. The free version of Kite, creatively named Kite Free, “has all the core features you’ve come to love.” This is not strictly true. Kite Free includes completions ranked by relevance, local code processing, documentation as you type, and function signatures as you type. The free version, however, used to also include Line-of-Code Completions. That feature, along with Multi-Line Completions, is now in the Pro version.
Kite has been free for over four years. Smith had told us in the past that monetization would come one day as an enterprise version. While a business-specific flavor is still in the works, the tool’s user growth meant it was time to start charging for features.
Not everyone has to pay for Kite Pro. University students can get Kite Pro for free by signing up with their school email address. The company plans to expand student access to high schoolers and others in the late summer, “just in time for the fall semester.”
Regardless of whether you’re going with Kite Free or Kite Pro, you’re getting an updated engine. Kite now uses deep learning, a type of machine learning.
“The field of using deep learning to understand code is evolving very quickly,” Smith explained. “The models that we are releasing today combine the state of the art from academia with our industry-leading code engine. Our models are trained on tens of millions of open source code files from a variety of code projects. The models are able to learn common patterns, along with the ability to understand the code that users are writing. As a result, the models can predict what users are going to — or should — type next. We provide these predictions through editors’ completions UIs so that users can focus on coding instead of typing, and save Google searches when they are having trouble remembering an API. We will stay at the forefront of ‘deep learning for code’ research so that our users always have a cutting-edge experience.”
Kite can provide completions “when editors like Visual Studio Code cannot understand the code,” Smith boasts. Kite’s completions even work alongside your IDE’s completions with filters designed to reduce noise.
Growth and next languages
In addition to the over 250,000 people coding with Kite every month, Smith shared some more numbers today. Kite’s main Git repo has over 30,000 code commits and its software totals 500,000 lines of code. The team has received over 6,000 pieces of written product feedback and has closed over 3,000 issues.
As for which programming language is next, Smith said the team has nothing to announce today. You can, however, go to Kite’s Let Me Know page to vote for the language you want next and sign up to get notified when it’s released.