After emerging from stealth in 2017, the San Francisco-based Harness now has a $1.7 billion valuation. The company’s customers include Kohl’s, McAfee, National Bank of Canada, UK Home Office, and Citigroup.

While Harness has a number of competitors, like GitLab and CircleCI, the DevOps market is expected to grow quickly: IDC projects that the market will reach $15 billion in 2023, up from $5.2 billion in 2018.

The new updates contribute to the company’s overarching aim of providing a unified pipeline for software delivery, sped up with intelligent features. The four new components of the Harness platform include Test Intelligence, Feature Flags, Cloud Auto-Stopper and Unified Pipeline.

The new Test Intelligence reduces test cycle time by up to 98% using AI/ML workflows to prioritize and optimize test execution. It automatically shifts the failed tests earlier in the build cycle so the developer will know, almost immediately, if the fix worked. Harness says it’s the only software delivery company on the market using test intelligence driven by AI and ML.

Feature Flags, meanwhile, allow developers to release new features without making them visible to users. With control over which pieces of functionality are on and off, outside of the code, developers can release, manage and govern features at scale. Companies can also test out more advanced and complex scenarios like feature A/B testing or software functionality variations, such as one- or two-step checkout.

With Cloud Auto-Stopper, Harness promises to reduce cloud compute waste associated with non-production environments (dev, QA, staging) by up to 75%. The new features follow Harness’ acquisition of Lightwing technology, which was integrated into the Harness Cloud Cost Management module. This allows engineering teams to quickly auto-stop and restart non-production environments.

Harness said its own engineering team was able to reduce AWS non-production spend by 25% and Azure non-production spend by 50% using the feature.

Lastly, Harness’ Unified Pipeline offers continuous integration and delivery, feature flags, cloud cost and change verification, all managed in a single pipeline.

What the release delivers is stable language features such as JVM records, sealed interfaces, online classes, and includes the new default JVM IR compiler.

More than 25,000 developers have already tried the new JVM IR compiler in IntelliJ IDEA.

The new compiler shares a unified pipeline and business logic with Kotlin/Native and Kotlin/JS IR compilers, which will allow developers to implement most features, optimizations, and bug fixes for all platforms simultaneously. It will also allow you to add custom processing and transformations that will automatically work on all platforms.

Auto-update to Kotlin 1.5.0

If you are a user of IntelliJ IDEA or Android Studio, it will give you the option to automatically update to the new Kotlin release as soon as it becomes available.

Stable language features

Kotlin 1.5.0 includes stable versions of the new language features presented for preview in 1.4.30:

• Support for JVM records to remain interoperable with the latest Java features – record classes. To use a Kotlin class as a record in Java code, make it a data class and mark it with the @JvmRecord annotation.
• Sealed interfaces. The sealed modifier works on interfaces the same way it works on classes.
• Sealed class improvements. Previously, all subclasses had to be in the same file with sealed classes.
• Online classes are a subset of value-based classes that only hold values.

Standard and test library improvements

The standard library provides many helpful features, including:

• The new stable APIs for unsigned integer types as well as for ranges, progressions, and functions for them.
• Extension functions for java.nio.file. Path to use the modern non-blocking Java IO in a Kotlin-idiomatic style.
• String and Chаr API improvements including the new locale-agnostic API for changing the case of strings and characters, a set of new functions for conversion between characters and their integer codes and numeric values, and multiplatform support for more char functions.
• Duration API changes including using a Long value for the internal representation and providing new properties for retrieving a duration as a Long value.

The test library comes with a set of important improvements:

• Single kotlin-test dependency in the common source set of multiplatform projects.
• Automatic choice of a testing framework for Kotlin/JVM source sets when you specify only the kotlin-test dependency.
• A number of assertion function updates.

Kotlin/JVM

Kotlin 1.5.0 brings the new JVM IR compiler, announced in Kotlin 1.4.0, becomes Stable and default. Compilation of Kotlin plain lambdas is Experimental, New default JVM target: 1.8. (the 1.6 target is deprecated), Improved handling of type nullability information from Java and Deprecation of @JvmDefault and old Xjvm-default modes.

Kotlin 1.5.0 provides support for compiler caches in debug mode for linuxX64 and iosArm64 targets. With compiler caches enabled, most debug compilations complete much faster, except for the first one. Kotlin 1.5.0 also provides deactivation of the built-in memory leak checker to avoid issues that may cause application crashes.