JACoB AI coding bot released as an open source project

Kevin Leneway, principal software engineer at Pioneer Square Labs (PSL), explores the potential of artificial intelligence in software development. This exploration led to the creation of JACoB, an acronym for “Just Another Coding Bot.” Originally launched earlier this year, JACoB has evolved through user feedback and validation. This week, PSL released JACoB as an open source project.

Opportunities and integration

JACoB integrates seamlessly with GitHub and other commonly used development tools. It can write and review code, convert Figma projects to code, automate routine tasks such as pull requests, and identify potential flaws and security bugs. The bot is designed to learn from patterns, which improves the quality of its results over time.

PSL emphasizes that JACoB includes built-in safeguards to prevent uncontrolled processes. Early adopters reported increased efficiency and productivity without compromising quality and safety. The tool mimics a workflow typically associated with junior developers that has been refined over decades.

Open source and customization

One of the standout features of JACoB is its open source nature. This allows developers to check the source code and contribute to improving it. Additionally, users have the option to run JACoB locally, providing additional data protection and code security. A hosted version is also available, free during the trial period.

Currently, JACoB is optimized for TypeScript and JavaScript. Leneway explained that the focus on a narrow use case was intentional to ensure high-quality results. The bot works particularly well with NextJS, Tailwind, and Figma for project-to-code transformation. In a YouTube video, Leneway demonstrated how JACoB works, showing how he creates and builds his own homepage.

While JACoB has not yet been spun off as a separate startup, Leneway suggested that could be the future. For now, our focus is on making the project open source and allowing developers to experiment with it. PSL wants to see how the community adopts and adapts JACoB.

Getting started with JACoB

Developers interested in JACoB can register on the project’s website or use the quick guide. The bot integrates directly with GitHub workflows, transforming Figma projects into deployable code and understanding the entire code base. Contributions are open to the project, and developers are encouraged to join the community to help shape the future of artificial intelligence in software development. Leneway posted a video on YouTube discussing the JACoB setup process.

Technical Documentation

The project includes a technical paper detailing the JACoB methodology and architecture. Evaluations of other design-to-code tools have shown impressive results. The bot is designed to be flexible and extensible, allowing for customization to meet specific needs while emphasizing privacy and security.

There was no standard test of how well design-to-code tools translated designs into actual code. Therefore, PSL built its own rating system to evaluate the capabilities of JACoB in practical situations. This test and evaluation ground is known as the JACoB Arena. During testing, JACoB stood out, beating out eight other popular tools and even human-written code. He achieved an ELO of 1183, which is over 100 points higher than his closest competitor. To learn more about how we achieved this, check out the technical documentation included in the repository.