JetBrains Launches Governance Platform to Prevent AI Agent Cost Sprawl

The Prague-based developer tools company unveiled JetBrains Central last week, a governance and execution platform designed to help organizations track return on investment, manage expenses, and coordinate AI agent workflows as they scale agentic AI deployments.

While deploying AI coding agents has become straightforward, enterprises struggle with visibility into their effectiveness and risk allowing them to become yet another unmanaged layer of technical debt. Oleg Koverznev, vice president and head of Agentic Platform at JetBrains, warned that the industry faces a replay of the cloud ROI crisis.

Everyone knows AI is a game changer. But it's very hard to prove that implementing these systems is making a difference in what the return is for the business.

Oleg Koverznev, JetBrains

When cloud adoption accelerated, organizations quickly faced pressure to justify spending, spurring the emergence of cost management and observability tools. JetBrains is positioning itself ahead of the curve by addressing similar pressures in the agentic AI space.

A January 2026 AI Pulse survey conducted by JetBrains among 11,000 developers revealed that 90% already use AI at work, and 66% of companies intend to adopt coding agents within the next 12 months. However, only 13% report deploying AI across the entire software development lifecycle. Koverznev noted in the launch announcement that code generation has become commoditized, shifting focus to a different challenge: "The real challenge is aligning outcomes with intent, along with managing the growing operational and economic complexity of agent-driven work."

The coordination problem nobody's talking about yet

Beyond ROI measurement, Koverznev highlighted a more urgent concern: what occurs when agent deployments scale beyond the point where organizations maintain visibility into their operations.

We envision that in the future there will be teams of humans and agent co-workers. The coordination across agents and humans is a big aspect — and this problem is emerging.

Oleg Koverznev, JetBrains

Most enterprises currently operate agents in isolated silos. Gartner projects that 40% of enterprise applications will incorporate AI agents by the end of 2026, compared to fewer than 5% today. At that scale, maintaining awareness of what's executing, its cost, and whether it achieved intended outcomes becomes a significant operational challenge.

JetBrains Central addresses this through a semantic layer that consolidates context from codebases, system architecture, runtime activity, and delivery infrastructure—enabling agents to operate with system-level comprehension rather than relying solely on prompt-level guidance.

To get predictable results, we need to give agents the necessary context and the ability to understand the code.

Oleg Koverznev, JetBrains

This capability integrates with Air Team, a collaborative environment for assigning work to agents and managing complex workflows. The platform connects with Slack, Atlassian products, and Linear. The underlying argument is that agent workflows operating independently from existing team infrastructure create redundant coordination layers.

An open platform plays in a market moving toward lock-in

While most AI platform vendors construct proprietary ecosystems, JetBrains is pursuing the opposite direction. Koverznev was direct about the strategy.

We don't want to own the stack. We want to provide the system so that customers choose which tools, which agents — and we give them the ability to control it.

Oleg Koverznev, JetBrains

The platform supports connections to any IDE or command-line interface, permits customers to supply their own API credentials for OpenAI or competing providers, and allows integration of external agents—including Claude, Codex, and Gemini CLI—through the Agent Communication Protocol without requiring custom development. An on-premises deployment option is in development.

We're standing on the shoulders of giants. LLMs provide great intelligence — we don't compete with that. We bring it all together into a controllable system.

Oleg Koverznev, JetBrains

A no-lock-in approach succeeds only through robust integrations, and this sector continues expanding rapidly. JetBrains is demonstrating commitment to its own platform—Hadi Hariri, senior vice president of Operations, stated: "We're increasingly leaning into agents and AI-driven workflows, which is creating a need for better visibility into costs and governance. That's why we've started piloting JetBrains Central internally."

Pricing: Fixed governance, variable execution

The pricing structure comprises two components: a fixed per-seat subscription for governance functionality covering both JetBrains and third-party users, plus consumption-based pricing for agentic execution. Koverznev outlined several pricing scenarios.

One developer can spend $100 a month. Another can orchestrate thousands of agents and spend $100,000. It's really possible.

Oleg Koverznev, JetBrains

The platform's purpose is to render spending transparent and connect it to measurable business outcomes including time-to-market and delivery costs.

The Early Access Program begins in Q2 2026 with a limited cohort of design partners evaluating JetBrains Central in production agentic workflows. The open-platform strategy depends on JetBrains reaching general availability before the market consolidates around a competing governance layer.

Source: The New Stack