Highlights:

  • Robin AI is the tool that Integral has released. It searches code files for potential problems using artificial intelligence.
  • With the help of the tool, developers can run automated software workflows that interact in various ways with their code repositories, such as by highlighting optimization opportunities.

Integral, a venture-backed data management startup, recently unveiled an open-source tool that can analyze the code of software teams and automatically highlight areas for improvement.

The tool is referred to as Robin AI. It searches code files for potential issues using artificial intelligence.

Integral, formally known as Promeed Inc., was established in 2022 and raised an undisclosed amount earlier this year. It is supported by a number of institutional investors, including publicly traded data management software provider LiveRamp Holdings Inc.

Integral offers a platform that aids healthcare organizations in making sure their data analytics projects adhere to legal requirements. The startup created Robin AI, the open-source tool it unveiled recently to increase the productivity of its software engineers. Integral claims that the number of bugs that need to be manually fixed by its developers has decreased by using the tool.

Co-founder and Chief Technology Officer of Integral, John Kuhn, said, “Because it serves as a trusty partner in enhancing the code development processes, we named the project after Batman’s sidekick Robin. It has served us well in boosting product velocity and reducing production bugs for Integral’s engineering team.”

When a developer submits a “pull request,” Robin AI gets activated. That’s a notification that a freshly written code file is prepared for deployment but must first be checked twice for potential problems. Following pull requests, it employs a machine learning model from OpenAI LP’s GPT series of neural networks to automatically examine code files for errors.

Some pieces of code must consistently function even when errors happen. Software developers must incorporate so-called error-handling mechanisms in order to reach the required level of reliability. Integral claims Robin AI can identify a code file’s absence of error-handling components and several other typical software flaws.

The tool also can detect subtler problems. For instance, it can identify code fragments with inadequately descriptive names. Developers find comprehending code with a name that describes its function simpler.

Robin AI produces a list of enhancement recommendations after identifying potential problems in a code file. It also offers code examples that demonstrate how those recommendations can be implemented. As an extra precaution, it generates a code quality score from 0 to 100, representing the issues’ severity.

Deployment is fairly straightforward because Robin AI is developed as a Docker container. It is made to function with the GitHub Action tool. With the help of the tool, developers can run automated software workflows that interact in various ways with their code repositories, such as by highlighting optimization opportunities.

The source code for Robin AI is available on GitHub. The tool is offered without charge under the MIT license.