CodeSee, a tech startup founded in 2019, has been on a mission to enhance programmers' understanding of the entire code base within companies. Their approach involved providing visual maps that illustrate connections between various services and systems, allowing developers to comprehend how changes might impact different links. In a recent announcement, CodeSee revealed a significant advancement to its platform — a generative AI capability, that empowers users to inquire about the code base in natural language.
According to Shanea Leven, the co-founder, and CEO of CodeSee, integrating generative AI aligns perfectly with their existing goals and use cases. While the fundamental mission remains unchanged, adding AI enhances insights into how people grasp and comprehend code.
The generative AI feature enables an interactive means of questioning the code directly within the Integrated Development Environment (IDE). For example, users can inquire about specific aspects, such as how logging functions or state management is utilized in the code base. In response, CodeSee provides a text-based answer accompanied by a code map, visually showcasing the connections between different elements of the code based on the question posed.
Leven explained the functionality further, stating, "You're able to kind of deep dive on where those different things are. And then from that question, we draw lines to where those files and functions are located in your IDE, so that you get the context of how something works." This feature serves as an intelligent extension, building upon CodeSee's existing capabilities.
CodeSee collaborated with tech giants Microsoft and OpenAI to develop this cutting-edge generative AI capability. The partnership proved beneficial as it allowed the company to rapidly introduce the new functionality without significantly expanding its team, which currently consists of only 11 employees.
Recently, CodeSee introduced service maps, offering companies an expansive view across multiple code bases. This comprehensive cross-repo visibility encompasses endpoints, message queues, Kafka pipelines, and various system components. By plugging into this feature, enterprise customers gain a high-level architecture view, providing them with a deeper understanding of their code base.
As this codebase understanding continues to expand its offerings, the introduction of an AI-based question engine is expected to further augment customers' code clarity. The new generative AI feature will initially be available in beta upon request. The company plans to collaborate closely with enterprise customers to fine-tune the tool before making it generally available later this year. This move showcases CodeSee's dedication to providing innovative solutions for programmers seeking to navigate complex code bases efficiently.