December 9, 2023

code view Founded in 2019, it aims to help programmers understand the entire code base inside the company. The idea is to provide a visual map so you can see the connections between services and systems and how changes affect different links. Today, the company announced that it is adding generative artificial intelligence capabilities to the platform, making it easy to ask questions about codebases in natural language.

According to Shanea Leven, the company’s co-founder and CEO, generative AI is a perfect fit for what they’ve been trying to do at CodeSee. “So our assumptions, our mission and our use cases don’t really change with the addition of AI, but when you add AI to the code, our insight into how people grasp and understand the code actually increases,” she told TechCrunch.

The Generative AI feature provides an interactive way to ask questions about code in the IDE.For example, you can ask questions like How logging works or how state management is used in your codebase, CodeSee provides text-based answers, complete with code diagrams where lines show connections between different parts of your codebase based on your question.

“That way you can drill down to see where these different things are. And then from that question, we actually draw the lines of where these files and functions actually are in the IDE so you get the context of how something actually works,” she said. It becomes a smart extension on top of what CodeSee already offers.

CodeSee AI, Generative AI question box on the left and code map on the right.

Image Source: code view

Leven said the company is working with Microsoft and OpenAI to help build new generative AI capabilities. This helps in many ways, as the company still has 11 employees, and the partnership allows them to quickly build new features without adding additional staff.

code recently viewed Launch service map Provides companies with visibility across multiple codebases. “It’s a full cross-repository visibility view of all endpoints, message queues, all Kafka pipelines, all things (in the system), plug that in and you get a full high-level architectural view that we didn’t have before, which really helps a lot of enterprise customers really understand what’s going on in their codebase,” Leven said.

As the company expands its offerings to give customers a clearer view of their code, having an AI-based problem engine should become even more useful. First, new generative AI capabilities will be available in beta upon request, she said. They hope to work with enterprise customers to refine the tool before rolling it out later this year.