JetBrains Central: 에이전트 소프트웨어 개발을 위한 개방형 시스템

Introducing JetBrains Central: An Open System for Agentic Software Development AI is beginning to change how software is produced. Instead of just assisting developers inside the editor, AI agents now investigate issues, generate code, run tests, and execute multi-step workflows. As this work scales, software development extends beyond individual tools or sessions. It becomes a distributed system of agents, environments, and workflows that operate across IDEs, CLIs, pipelines, and collaboration tools. In this new model, code generation is cheap and no longer a bottleneck. The real challenge is aligning outcomes with intent, along with managing the growing operational and economic complexity of agent-driven work. Without control over these factors, systems become harder to reason about, scale, and sustain. This shift is happening quickly. Of the 11,000 developers worldwide who responded to the January 2026 JetBrains AI Pulse survey, 90% already use AI at work. Adoption of coding agents is also accelerating – 22% of developers already use AI coding agents, and 66% of all companies surveyed plan to adopt them within the next 12 months. However, most of AI’s impact remains limited to individual productivity. No more than 13% of developers report using AI across the entire software development lifecycle, such as for code review or in the release pipeline, and organizations struggle to translate AI use into measurable improvements in software delivery speed, system reliability, or cost efficiency. JetBrains Central: The control and execution plane for agent-driven software production JetBrains Central transforms discrete AI-powered workflows into a unified production system. It connects tools, agents, and infrastructure, allowing automated work to run, be monitored, and be managed across teams – with clear visibility into results, costs, and performance. Developers can initiate and manage agent workflows from the tools they already use – JetBrains IDEs, third-party IDEs, CLI tools, web interfaces, or integrations. Agents can come from JetBrains or external ecosystems, including Claude Agent, Codex, Gemini CLI, or custom-built solutions. JetBrains Central provides three core capabilities: - Governance and control Policy enforcement, identity and access management, observability, auditability, and cost attribution for agent-driven work. Some of these functionalities are already available via the JetBrains Central Console. - Agent execution infrastructure Cloud agent runtimes and computation provisioning that allow agents to run reliably across development environments. - Agent optimization and context Shared semantic context across repositories and projects, enabling agents to access relevant knowledge, and task routing to the most appropriate models or tools. JetBrains Central is not a monolithic platform. Instead, it functions as a layered system that connects developer tools, AI agents, and development infrastructure. This architecture enables a no-lock-in approach to AI-driven development, allowing organizations to integrate new tools and models while preserving and extending the systems they have already invested in. This eliminates the need for costly replatforming. Context, semantics, and integrations across the software delivery system To be effective, AI agents must operate within real software production systems and organizational contexts – not in isolation. JetBrains Central connects agents directly to the systems where software is built and run, including repositories, knowledge bases, delivery pipelines, and infrastructure. This allows agents to execute work within existing development workflows, rather than in separate AI environments. At the core of this system, we are building a semantic layer that continuously aggregates and structures information from code, architecture, runtime behavior, and organizational knowledge. This enables agents to move beyond prompt-level interactions and operate with a system-level understanding of how software is designed, how it behaves in production, and what outcomes are expected. On top of this foundation, JetBrains Central provides intelligent routing and task optimization, selecting the most appropriate models, tools, and execution paths for different tasks. Agents collaborate with human teammates through the tools teams already use – such as Slack, Atlassian products, or Linear – ensuring that agent-driven workflows remain integrated into existing development systems instead of becoming isolated AI workflows. Coordinating human and agent workflows with JetBrains Air The recently launched Air App provides a dedicated workspace where developers can organize tasks, run agent-assisted workflows, and review results while staying close to their development environments. For teams, JetBrains is developing Air Team – a space for coordinating work between humans and agents, enabling teams to organize tasks, run multi-step workflows, and stay aligned as work happens acro