Newly-funded AI startups are rethinking the code editor

Since the public launch of OpenAI's large language models last year, AI has found its way into many of the tools developers use in their daily work. But one ecosystem has seen a surge in popularity: companies bringing AI into code editors.

Many of them have raised venture funding, most often in pre-seed and seed rounds. A few, such as Mutable AI and TabNine, have gone on to raise Series A or Series B funding. The biggest, like Sourcegraph, HuggingFace, and Zhipu AI, have raised hundreds of millions, although code editor extensions represent just one product within their wider platforms. All together, these companies have brought in at least $800m from investors (even excluding a handful that have not disclosed their funding amounts). What is driving their sudden rise?

Two concurrent trends shed light on why these startups are thriving. The first and most obvious is that AI-enabled development is becoming more widely accepted. In the latest Stack Overflow survey, 70% of developers said they are using or are planning to use AI tools in their development process this year. Last year's survey didn't include a single question about AI tooling — a testament to how much has changed since then.

The second trend is that modern code editors and IDEs make it easier than ever for developers to find and try new tools. Visual Studio Code, used by nearly 74% of developers, helps them tap into tens of thousands of extensions directly from their editor. JetBrains, creator of popular IDEs like IntelliJ and PyCharm, maintains an official marketplace with upwards of seven thousand plugins. Extensions like Bito, Codeium, and Safurai can write and edit code with the latest AI models soon after being installed.

How long can companies sustain the current AI boom? Overall fewer startups are making it from priced seed rounds to Series A. Peter Walker, Head of Insights at Carta, says that "most of those that raised Seed capital in 2021 were planning on raising a Series A this year — and that capital just hasn't materialized." AI development tools will have to buck this trend in the coming months. With enough momentum, they might just be able to do so.