“I feel the cost [of Codacy] is justified for multiple reasons. First, it’s a great tool that provides automated linting for a large number of languages and technologies. Second, it’s highly customizable. Third, it needs minimal setup and “just works” for most engineers without even leaving the PR screen on GitHub.”

About Leia Inc.

Leia Inc. (“Leia”) vision is to change the way people connect, create, educate and learn by removing technology barriers. Based on a breakthrough in nano design and manufacturing, Leia’s development of Lightfield technology, hardware, software, optics, and firmware all come together to transform existing displays into an explosion of beauty and emotion. Headquartered in Menlo Park, California and with worldwide offices spread across California, New York, and China, Leia has dedicated more than 50% of their workforce to engineering roles.

Primary languages and frameworks used by their development team include: C++, Java, Python, Unity3D, Android, OpenGL.

Challenge

Leia needed a code review tool with automatic setup that could work across the different programming languages, technologies and tools used in their engineering department. The tool would help ensure that codebase follows a set of standardized best practices, making it easier for engineers to dig into each other’s code for triaging and collaboration.

Why Codacy

Leia chose Codacy for its practicality and ease of use. Codacy easily integrates with Leia’s GitHub workflow and needs no manual set-up. Other automated code review products on the market took longer or more effort to integrate or required the use of external tools outside GitHub. Although the Leia team previously considered building a tool using existing CI/CD (Continuous Integration/Continuous Delivery) pipelines, they realized that with Codacy there was no need.

With Codacy

Codacy has become part of the engineering team workflow and overall culture at Leia. When engineers integrate a new repository, Codacy automatically checks which programming languages are used in the repository and sets everything up for code review on its own with no human guidance.

Codacy checks code prior to peer review and points out issues before a peer reviewer looks at the pull request. This saves time and helps the team follow the same set of practices and linting rules, making collaboration much easier.

Leia uses Codacy’s existing rules setup for their repositories or enables Codacy’s “best practice” rules for each language. There are rare cases, such as with Android or Unity3D technologies (which Java and C# linters do not cover 100 percent of the codebase) in which Leia’s needs extend beyond Codacy’s default settings. For these, Leia’s team is in regular contact with Codacy for support. Manual interventions are only warranted with specific additions or changes to the review process.

Other Benefits

Initial, automated review enables focus on code logic: By pointing out “nit-pick” type issues (like missing white space) in its first pass review, Codacy allows reviewers to focus on code logic, rather than on language-specific formatting or linting issues. This is particularly helpful when code reviewers are less proficient in the coding language(s) used by code authors.

Builds trust in tools: The team trusts Codacy to autonomously perform effective code review and “can be sure that all aspects of the codebase are being covered” according to Puneet Kohli, a Software Engineering Manager and former Senior Engineer at Leia.

Boosts engineering team confidence: Team members know that regardless of coding language, their code has been sanitized and the most error-prone issues like potential memory leaks are covered. In one instance, this helped a team of domain experts working on a Digital Avatar project, provide high-quality reviews of technologies that they were not experts on “without the fear of the code in question being of poor quality and not identifying it.”

Increased awareness to address issues: Codacy detects issues that the team was not initially aware of. This includes potential memory issues, especially in C++ codebases, and linting and formatting issues for README files, which have now been made consistent across projects.

Invest and maximize talent with learning: Over time using Codacy, Leia’s engineers learn to fix issues sooner (while writing code). As a result, Codacy flags less errors than it did when Leia initially started using the tool. Due to this, Puneet Kohli, who initially introduced Codacy to Leia while he was Senior Engineer, calls Codacy an “investment” in the “future of engineering talent.”

Integrated into engineering culture to provide anxiety relief: Codacy helps quell anxiety of new developers, particularly recent graduates, joining Leia. Responding to newcomers worried about producing code to meet existing company coding standards, Puneet advises “they should not worry about it, and simply submit a pull request, and watch the magic happen.”

Future

Since Leia began using Codacy, the team has roughly doubled in size. The company expects the rapid growth to continue as they scale the engineering organization. It aims to continue integrating Codacy into projects which are peer-reviewed.