Mike Vizard examines how Harness is extending its AI platform with new modules designed to automate code maintenance and streamline DevOps workflows for development teams.

Harness Adds New AI Modules to Automate DevOps Pipelines and Maintenance

Author: Mike Vizard

Harness has introduced two new modules to its artificial intelligence (AI) platform for DevOps, aiming to enhance automation in code maintenance and facilitate deployment rollbacks. These additions are part of a broader move to apply AI-driven solutions to routine DevOps challenges, enabling engineers to focus on higher-value work.

Key Additions to the Harness AI Platform

Autonomous Code Maintenance (ACM):
- Enables developers to describe tasks in plain English.
- Automatically creates builds that pass security and functional tests.
- On build failure, the AI diagnoses issues and suggests or applies remediations by creating pull requests until the problem is resolved.
- Identifies and removes stale feature flags, opening approval requests for review.
AI Verification and Rollback Module:
- Integrates with observability platforms to collect deployment metrics and log queries.
- Builds health verification profiles and automates rollback to the last successful deployment if issues are found.
Architect Mode:
- Engages DevOps engineers interactively to design pipelines based on organizational practices for security, quality, and compliance.
Release Orchestration Updates:
- Uses natural language processing to model and codify new release workflows, translating instructions into structured YAML code.
AI Agent in Developer Portal:
- Provides DevOps teams with AI-driven support within the internal developer portal (IDP).

Impact on DevOps Teams

According to Harness Field CTO Nick Durkin, the adoption of these AI-driven capabilities aims to reduce manual, repetitive work for DevOps engineers—enabling them to concentrate on complex, value-added tasks. The increased automation lowers the barrier to entry for new developers, providing opportunities for faster growth and less reliance on deep DevOps specialization.

Durkin notes that DevOps teams, on average, still spend between 60% and 70% of their time on routine tasks. The integration of AI in these workflows is expected to reduce burnout and foster more efficient and scalable application delivery.

Early Adoption and Takeaways

While it is still early in the integration of AI into DevOps, these enhancements from Harness suggest a trend towards more autonomous, resilient, and developer-friendly operations. AI capabilities in pipeline design, health checks, and remediation highlight the platform’s commitment to security, reliability, and compliance in software delivery.

For more details, read the original article by Mike Vizard at devops.com.

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