Amazon CodeGuru customers

Amazon CodeGuru helps Cognizant development teams deliver mission critical software for our customer's digital transformation programs. Incorporating CodeGuru in our development workflows improves and automates code reviews, helps our DevOps teams proactively identify and fix functional and non-functional issues and ensures that the deployments exceeds the performance, security and compliance requirements of our customers across industries and regions.

A key component of our future roadmap is to turn all our products into cloud-native products that leverage the incredible array of managed services available at AWS. Rebuilding old school, on-prem architectures, and transforming them for the cloud brings a whole set of engineering challenges that range from keeping abreast with all the latest services to adjusting to the paradigm shift that is associated with these architectures. CodeGuru is an incredibly valuable tool that helps optimize our products’ performance while making sure that we are leveraging these services with all the best practices in place. Without tools like CodeGuru Reviewer, we wouldn't have been able to rewrite entire products like FogBugz to be AWS cloud-native. We are now using CodeGuru Profiler to optimize a number of products including EngineYard's container-based ‘No Ops’ platform and well as the next generation of the Jive collaboration platform.

Amazon CodeGuru Reviewer has helped us create a feedback loop for our developers and improved the overall quality of our code. The new pricing model makes it much easier for us to budget and operationalize adding it to our software development processes. We can now scale our code reviews based on our project size and deployment schedules.

Part of application development is the creation of performant systems as well as the feedback and continuous improvement of existing systems. This starts with a strong architectural foundation but often ends in the details of the application code. When our team at EagleDream is digging into these details there are a variety of tools at our disposal, and using both static and dynamic analysis is helpful. We’ve found that the runtime analysis offered by the CodeGuru Profiler is one of the simplest and fastest to get running, and generates insights into the application code that are easy to remediate. Using this tool we’ve been able to quickly hone in on portions of an application that represent bottlenecks that would have otherwise been difficult to spot, then develop changes that can be implemented and tested with a very fast feedback loop. This allows us to continuously deliver and improve our own workloads and, the workloads of our customers, making them more performant while saving on cost at the same time.

With CodeGuru, we have built automated code reviews directly into our pipelines, which means my team can deploy code faster and with more confidence. We use CodeGuru Reviewer’s recommendations based on ML and automated reasoning, to focus on fixing and improving the code, instead of manually finding flaws. The addition of Python has made CodeGuru even more accessible for us.

Poor quality code adds complexity to the system and can become technical debt at some point. On the other hand, as long as consistent code quality is maintained, scaling the system won’t prevent developers from extending features as the code itself is simple. At RENGA, the code review process is important, however, it should not increase workload for reviewers or become a bottleneck in development. Powered by machine learning, Amazon CodeGuru Reviewer helped us automate code reviews and reduced the workload required on reviewers. We could seamlessly integrate Amazon CodeGuru Reviewer into our existing development pipeline. Furthermore, learning the best practices of coding – which we were not aware of – has helped us develop with more confidence.

Our use of Amazon CodeGuru Profiler is very simple but extremely valuable. We've optimized our worst performant service to reduce its latency by 15% for the 95th percentile in a typical work day.