The Code Guru service offers two tools, a bug scanner that developers can run while reviewing code called the CodeGuru Reviewer and Code Guru Profiler that can identify code responsible for CPU over-utilization.
A preview version of CodeGuru was launched by Amazon last December and presented the CodeGuru service as a useful tool for programmers to automate code reviews. With Code Guru, developers can uncover bugs, find approaches to solve them, and receive other valuable insights to improve their code before the product reaches users.
CodeGuru comes with a 90-day free trial, after which Amazon will charge you $0.50 per 100 lines of code from your source code repository.
Developers who run their apps on AWS compute instances can reduce their computing costs with a much more efficient code thanks to the new service. CodeGuru is an AI-based tool that consists of machine learning models trained on over 10,000 open sources GitHub projects, including Amazon’s code reviews. According to AWS, this makes CodeGuru highly accurate in its analysis.
To use the tool, all developers have to do is add AWS CodeGuru as one of the code reviewers and commit their code to GitHub as they do regularly. This will open a pull request for vetting the code. It also comes with a dashboard that displays information about the code being reviewed, including the number of lines analyzed, the number of recommendations generated, and the overall status. Developers can go through the recommendations and accept or reject them.
Amazon claims that CodeGuru has been used in the making of over 30,000 of their production applications and helped them save millions of dollars in computing costs. The tool has even been used in preparation for Prime Day to give a boost of performance.
CodeGuru Features
The profile provided by CodeGuru is focused on helping you save costs and even provides the cost of inefficiencies along with how you can reduce these costs while improving performance.
One user of CodeGuru is the popular software management company Atlassian. The head of engineering at Atlassian, Zak Islam, has stated that CodeGuru helped them reduce code reviewing times to a few hours or even minutes instead of a few days in case of anomalies. This has left their developers free to focus on meeting deliverables and business goals instead of puzzling over problems in the code.
Conclusion
The VP of Amazon Machine Learning has highlighted the importance of CodeGuru for customers who develop large applications with millions of lines of code. Having inefficiencies in every few lines of code can be detrimental to such companies and developers in terms of costs and time & resource investment. An AI-powered tool with a high degree of accuracy, such as CodeGuru, can be a game-changer for the industry. CodeGuru may prove to be an indispensable service for developers and companies all over the world. For now, the new service is available in selected areas of the US, European Union, and the Asia Pacific. AWS plans to expand this availability and will be announcing CodeGuru for other regions in the upcoming months.