The characteristic flag firm LaunchDarkly has introduced various new additions to its platform that can assist builders “ship on the velocity of now.” These bulletins have been unveiled at LaunchDarkly’s annual person convention, Galaxy ‘24, occurring at this time and tomorrow in San Francisco.
First, the corporate introduced a number of person expertise updates, equivalent to improved navigation, multi-environment views, dashboard shortcuts, and the flexibility to trace modifications for all assets.
Moreover, to allow smoother releases, the corporate introduced Launch Assistant, which helps standardize and automate features of the software program launch course of. It permits customers to create and handle their launch pipelines, outline audiences for every launch section, management approval and necessities, and automate rollouts.
It additionally introduced a brand new LaunchDarkly CLI the place builders can work of their improvement environments, documentation, and GUI with out switching between them, and an extension for GitHub Copilot that enables customers to create flags, get summaries about them, and examine the standing of flag cleanup.
Subsequent, the corporate introduced Launch Guardian, which permits groups to determine and repair software program points earlier than they make it into manufacturing. It integrates with fashionable monitoring and safety instruments, equivalent to Sentry, Section, and OpenTelemetry, and permits groups to correlate points with particular releases.
“Releasing software program includes a fragile stability between velocity and threat mitigation. Groups usually should select between velocity and security. With LaunchDarkly’s Launch Guardian, you now not should make this trade-off. Launch Guardian lets you monitor the impression of your releases in real-time, making certain swift subject detection and mitigation,” Claire Vo, chief product officer at LaunchDarkly, wrote in a weblog submit.
One other new characteristic is Superior Experimentation, which permits improvement groups to measure the enterprise impression of options and releases and conduct experiments to check potential new options.
“What good is it to ship options constantly in the event that they fail to have the supposed enterprise impression? That’s why LaunchDarkly combines characteristic administration and experimentation right into a unified answer and empowers groups to combine experimentation proper into their launch course of,” Vo wrote.
And at last, the corporate unveiled new AI options equivalent to AI Immediate and Mannequin Templates, an AI Variation Builder, and a GenAI Experiment Builder.
“AI has turn out to be a prime precedence for improvement groups. However AI improvement comes with a bunch of challenges: course of failures, mannequin conflicts, a scarcity of efficiency measurement, and many others. Luckily, with LaunchDarkly, groups acquire runtime management over configurations, iterate with stay person suggestions, and seamlessly combine generative AI flag templates to make sure assured, measurable, and impactful AI releases—all inside a single unified workflow,” Vo wrote.
