Greptile raises $4 million to create AI-powered codebase expert

Greptile raises  million to create AI-powered codebase expert

When you look at how generative AI is implemented in developer tools, the primary focus is on code generation, as is the case with GitHub Copilot. Greptile, an early-stage startup founded by a group of Georgia Tech graduates, decided to take a different approach: using artificial intelligence to help developers understand the code base.

Instead of using just a chat interface, as most vendors did, the startup delivered one other unique twist by building an API that developers could connect to the codebase and create custom applications based on AI-powered queries. On Thursday, the company announced a $4 million seed round.

- Advertisement -

Graptyl CEO and co-founder Daksh Gupta says the Greptile bot is like having a highly experienced colleague who understands your code in-depth. “That’s why we build AI tools that understand large code bases in companies, because over time, when multiple developers are working on a code base, it becomes very difficult to understand it,” Gupta told TechCrunch.

“The API principally has two parameters: one is to connect the repositories you would like to reference and make sure they are indexed. Once the repositories are indexed by the system, a natural language query is added akin to: “How does authentication work in this codebase?” – he said.

The startup was launched in July last yr after the founders got here up with the idea for the company during a hackathon. They launched the product, quickly grew to about 100 customers paying $10 to $20 a month, applied and got into Y Combinator for the 2024 winter batch.

However, before founding Greptile, while still in college, the founders created a chatbot for enterprise feedback management and raised enough money to move to San Francisco, where they quickly discovered that there was a large gap in their knowledge when it got here to starting a company.

“First, we made the canonical mistake of focusing on attracting investors slightly than real customer needs, building things that were probably good startup ideas as an alternative of looking for real problems that real people have and solving them with technology. We did all these items. And we learned the hard way that we shouldn’t do that,” he said.

Once they modified their focus to something more serious, the pieces began to fall into place. “No one told us that what you are doing is the purest type of capitalism. You create value for people to the point that they are willing to pay you for the value you create for them.

Y Combinator has learned its lesson about customer centricity. Today they have 500 paying customers, including individuals, teams and organizations.

As the company grows, founders have greater goals. “We want to provide software teams with the building blocks they need to create custom, tailored AI developer tools tailored to their system configuration,” Gupta said.

Today’s round was led by Initialized Capital with participation from various distinguished industry angels.

Latest Posts

Advertisement

More from this stream

Recomended