Mike Vizard covers how Harness has launched an AI-powered DevOps platform that uses knowledge graph–driven agents to automate a wide variety of tasks across the software delivery lifecycle, aimed at reducing manual effort and streamlining DevOps workflows.

Harness AI-Powered DevOps Platform Launches to Automate Software Delivery

Author: Mike Vizard

Harness has announced the general availability of its AI-powered DevOps platform, designed to transform how DevOps teams manage and automate tasks within the software delivery lifecycle (SDLC). The new platform employs knowledge graph–driven AI agents capable of automating pipelines, testing, and diverse operations.

Key Features of the Harness AI Platform

  • Knowledge Graph Foundation: The platform is underpinned by a Software Delivery Knowledge Graph, continuously updated with data from Harness and third-party DevOps tools. This enables AI agents to:
    • Generate and optimize pipelines
    • Rollback deployments automatically
    • Perform root-cause analysis on failures and incidents
  • Natural Language Interface: DevOps teams can describe their pipeline requirements in natural language, allowing the AI to build and deploy pipelines aligned with organizational policies and guidelines.
  • Automation Scope: The AI-powered platform can:
    • Create and maintain tests and chaos experiments
    • Detect vulnerabilities across environments
    • Surface actionable insights for cloud cost optimization
  • Security and Data Privacy: Data collected by the Harness AI platform is not used to train the underlying AI models; it is strictly for operational analysis and decision automation.

Impact on DevOps Workflows

Harness reports that beta users saw significant improvements:

  • Downtime reduction: Customers cut downtime in half
  • Faster debugging: Time spent on pipeline debugging was reduced by 50%
  • Accelerated testing: Test cycle times dropped by 80%, while test maintenance efforts decreased by 70%

Addressing DevOps at Scale

The platform targets common DevOps pain points—manual scripting, brittle workflows, and coordination overhead—by assigning repetitive and complex tasks to AI agents. This automation allows engineers to focus on oversight and higher-level orchestration. As more AI agents are incorporated, DevOps teams will transition towards managing platforms that orchestrate both human and AI contributors.

The Future of DevOps Automation

With the adoption of AI agents in software delivery, tasks that once contributed to burnout and operational drag can now be streamlined. The Harness approach means that organizational work on platforms, policy adherence, and delivery optimization becomes a key part of orchestrating large-scale, adaptive DevOps practices.

Related:

Learn more at: devops.com

This post appeared first on “DevOps Blog”. Read the entire article here