Harness today previewed a revamped user interface for its platform for managing software deployments that makes it simpler for software engineers to manage teams of artificial intelligence (AI) agents using a forthcoming Harness Software Factory platform.

Speaking at an {unscripted} NYC 2026 event, Harness CEO Jyoti Bansal told conference attendees the Harness Software Factory will make it simpler for DevOps teams to enforce standards across agentic AI engineering workflows. Rather than allowing AI agents to reinvent the way an application is developed every time, the Harness Software Factory makes use of specifications to enforce standards in a way that both applies controls and ultimately reduces the number of tokens that might otherwise be consumed.

Harness also previewed Vibe Mode for Harness, an interface for the Harness platform that applies policies to code being developed by so-called citizen developers who are generating code using AI tools.

Finally, Harness revealed a Flex Pricing option through which DevOps teams are able to purchase a pool of credits that can be applied across units of the Harness platform rather than having to license each module separately.

The Harness platform now revolves around four core AI agents that manage software delivery, security testing, runtime security and cost management. Each of those agents then assign specific tasks to sub-agents to autonomously complete a task.

At the core of those AI agents is a knowledge graph Harness developed to enable AI agents to manage DevOps workflows at machine speed, noted Bansal. Currently, the Harness knowledge graph is able to track 567 different types of entities, with 29 billion context updates now being made per month.

The overall goal is to eliminate the need for software engineers to write and manage scripts to construct a DevOps workflow, noted Bansal. That’s critical because existing DevOps workflows designed for humans will not scale in the AI era when application developers may soon be creating 15 to 20 pull requests (PRs) a week, he added. The controls required to safely deploy applications can now be managed by a set of AI agents, said Bansal.

Those AI agents are required because AI coding tools lack the context required to successfully deploy code in a production environment, he added. In essence, there will always be a need to separate areas of engineering concerns even as DevOps workflows are extended into everything from security and compliance to database management.

It’s not clear at what pace DevOps teams are injecting AI agents into DevOps workflows, but Bansal said in time there will be multiple levels of automation, ranging from fully autonomous to augmenting a human engineer. In most cases, organizations, depending on the nature of the application, will apply multiple levels of automations across different use cases, noted Bansal.

As AI agents become more embedded into DevOps workflows, programmatically managing the development and deployment of custom applications is only going to become more accessible to a broader number of organizations. The challenge and the opportunity now is determining how best to re-engineer those workflows in a way that ultimately leads to better applications being deployed faster than ever.