Harness introduces AI code and DevOps agents to accelerate its software delivery platform

Harness Inc., the software delivery startup that provides developers with tools to update and monitor their applications on the fly, has kicked off its annual user conference with a major update.

The company’s comprehensive software delivery-as-a-service platform is being expanded with a multi-agent artificial intelligence architecture that promises to transform developer workflows, improve developer experiences and, most importantly, increase productivity.

The latest release, announced at Harness’ 2024 {unscripted} event, is said to integrate AI capabilities into the core of the platform, creating an ecosystem of new AI-powered assistants that can help developers with virtually every aspect of software delivery.

The Harness platform already makes extensive use of AI, deploying machine learning algorithms to monitor new software releases in case they pose problems for developers. The service makes it easier for users to understand what’s happening in the core environment of an application. If a new update starts acting strangely, they can initiate an automatic rollback to the previous version of that app.

The platform is designed to support, among other aspects of software delivery, continuous integration and continuous deployment, or CI/CD, methodologies, a practice in which companies release frequent updates to their software and applications, often multiple times per day. Historically, developers would release new features to their software only once every few weeks or months.

With today’s update, Harness adds new AI capabilities. This time, however, it’s not about monitoring software releases, but specifically about supporting developers with the various tasks they need to perform to roll out those updates.

The new features include specialized AI assistants, such as the AI ​​DevOps Assistant, which is intended to support developers throughout the software development process. It does this by automating routine pipeline generation, optimizing workflows, tracking deployments and recommending improvements, and diagnosing and repairing any software issues.

There’s also a new AI Code Assistant, which is similar to GitHub’s Copilot tool, that helps developers write their software updates by suggesting code, testing it for issues, and so on. It’s all about helping developers speed up the coding process and ensure consistent quality. Meanwhile, there’s an AI Productivity Insights tool that’s supposed to help by measuring and optimizing the impact of the new AI assistants.

Additionally, Harness announced a new Database DevOps feature that simplifies the deployment and management of database changes by integrating them into users’ CI/CD pipelines and automating policy enforcement.

In an interview with SiliconANGLE, Harness founder and Chief Executive Jyoti Bansal (pictured) said the AI ​​DevOps Assistant and Database DevOps capabilities are entirely new innovations that no one else can offer. “We’ll have more of these assistants over time,” he added.

The main goal of Harness is to make it easy for developers to adopt and implement a CI/CD workflow. Before Harness, CI/CD was always a major pain for developer teams, with a complex set of open-source tools that would take months to set up. With Harness, developers can easily implement a pre-packaged set of CI/CD features in a matter of minutes.

Bansal said the company has come a long way since launching its first continuous software delivery product in 2018, expanding its platform to offer the most comprehensive set of CI/CD tools available. After starting with continuous delivery, Harness added continuous integration tools, followed by feature flagging capabilities and, most recently, cloud cost management tools.

“We see ourselves as an end-to-end software delivery platform,” he said. “We’ve systematically taken one piece of the software delivery process at a time and added it to our platform.”

The company is now accelerating that drive to cover the entire software delivery process, as there were many more new features in today’s release. For example, it also introduced “cloud development environments,” which are pre-configured, remote, and secure cloud-based environments for writing code and debugging, plus new supply chain security features and a new artifact registry.

Finally, the company said it is making a fully open-source version of its platform available to developers, allowing them to code, build and manage artifacts and deploy software in one centralized environment for free.

“My goal is to build a large platform company,” Bansal said. He wants Harness to be something like Salesforce Inc. or Workday Inc., but with a focus on software engineering. “Today, it’s all very disjointed,” he added.

With reporting by Robert Hof

Photo: Harness