Rollbar Inc. today updated its Continuous Code Improvement Platform with a new “automation-grade grouping engine” that it says can eliminate the daily noise and manual work involved in fixing broken software code, so developers can focus more of their time on building and updating their applications.

Rollbar, whose platform provides “monitoring software-as-a-service” capabilities for software developers, said today its new grouping engine relies on machine learning to determine patterns in code that can then be used to identify errors made in real-time.

The engine is a big improvement over its previous, hard-coded grouping engine, which failed to keep up with the constantly changing error types found in code. By regularly running its machine learning models against against new errors from its customers, Rollbar said it can recognize any new error types that emerge and group them more effectively.

The update comes at a time of strong growth for Rollbar, which said today it has exceeded 5,000 paying customers for the first time. That growth follows an $11 million Series B round of funding announced in March.

With the improvements to its grouping engine, Rollbar said it has created three new artificial intelligence-assisted workflows that will ensure developers spend less of their time on monitoring, investigating and debugging problems.

For example, the new Rollbar Toolkit for Kubernetes integrates the open-source container orchestration software Prometheus and Weaveworks Flagger, giving developers the ability to halt, sustain or expand progressive deployments automatically using Rollbar’s error data. That means developers no longer need to monitor their deployments for errors manually, so they can respond quickly when something does show up, Rollbar said.

Automated Feature Flag Triggers meanwhile, makes it possible to proactively discover errors and automatically switch off any feature they might effect, based on new and reactivated errors associated with that feature flag. The benefit is that it helps to minimize any impact on users when something does go wrong, according to Rollbar.

Finally, the Automated Issue Tracking workflow makes it possible to automatically generate tickets in issue tracking tools such as Jira or GitHub for any new errors that show up, based on a predetermined set of rules. This helps to ensure that all issues will be addressed, without developers needing to create a ticket themselves manually, Rollbar said.

“Our AI-assisted workflows, powered by our automation-grade grouping, are an industry first to reduce the daily noise and manual work of fixing code for developers so they can focus on building and innovating,” Rollbar co-founder and Chief Executive Officer Brian Rue said in a statement.

Image: Rollbar/Facebook