Agentic AI AI JetBrains New Products News Releases
AI can create code. Organizations motionless have to create software. Agentic betterment is changing how application gets made, but it hasn’t changed what it expenses to be wrong.
Six months ago, we began publically experimenting alongside agentic betterment environments. Around the identical time, we introduced JetBrains Central as an open authority and implementation scheme for agent-driven development. We subsequently began rolling out JetBrains Central CLI, shared context, haze agents, automations, governance, and AI disbursal controls for teams and organizations.
Today, we are bringing this activity together as JetBrains Air: an open, coherent scheme of products for developers, teams, and organizations, inner and beyond JetBrains IDEs. It is multi-surface and multi-service. Each merchandise solves a distinct problem, but the products activity improved together.
JetBrains Air marks a important enlargement in what JetBrains is construction for. For 26 years, we have focused chiefly on the idiosyncratic developer workbench. Now, we are construction for the wider scheme through which agentic activity is initiated, executed, coordinated, reviewed, and governed.
The IDE remains crucial to JetBrains’ future. The era in which the entire application betterment scheme can be contained in one opening is ending. As part of our continued investment, we are now bringing the foundational agentic cognition into JetBrains IDEs, giving expert developers an surroundings anywhere they can activity efficiently alongside agents during understanding, changing, and verifying the resulting code. JetBrains Air connects the wider scheme evolving about it.
That scheme is according to a center condemnation that the forthcoming of agentic betterment volition be multi-vendor. No sole model, agent, or assistance volition be correct for all developer, team, or task.
From one merchandise to an open scheme of products
This strategic change has a applicable consequence: JetBrains Air cannot be fair another delegate or betterment environment. It must nexus products for idiosyncratic work, squad coordination, organizational control, context, and procedure automation – and remain open to the tools and agents developers choose, including those JetBrains does not build.
JetBrains Air includes products that are accessible today alongside others that volition be introduced as the scheme develops:
- Air in JetBrains IDEs – a complete agentic betterment cognition for directing and orchestrating agents and verifying their activity inner JetBrains IDEs.
- Air Teams – a new way to coordinate and automate software-delivery workflows involving developers and autonomous agents.
- Air Governance (formerly JetBrains Central) – organizational policy, visibility, auditability, disbursal management, and accountability for AI-assisted and agent-driven development.
Junie is JetBrains’ coding delegate for expert application development. It volition be supported throughout all Air surfaces.
Air in JetBrains IDEs gives developers the surroundings to straightforward agents and verify their output using JetBrains’ code intelligence. Air Teams turns idiosyncratic delegate action into coordinated squad workflows. Air Governance makes that action visible, governable, and accountable throughout the organization.
But an open scheme cannot halt at JetBrains’ own products. The Agent Client Protocol (ACP) standardizes the association between an IDE and an agent’s complete harness, including its planning, logic, tools, example routing, and observability. Through the ACP Registry, developers can detect and run a expanding range of compatible agents during continuing to activity inner JetBrains IDEs.
Air Governance is designed to broaden visibility and disbursal governance throughout providers and the distinct tools through which agentic activity takes place. This method developers can choose the agent, model, or assistance suited to the project without forcing the institution to provision up context, visibility, or control.
Together, the Air products authorize activity to move between developers, agents, tools, and environments without losing the environment and controls surrounding it.
Individual acceptance has moved faster than organizational infrastructure
Since March, our products have progressed significantly, but so has our understanding of what agentic betterment requires.
Developers have been adopting agents faster than organizations can build the infrastructure about them. Agent capabilities have advanced, and distinct models and agents have proven helpful for distinct tasks. However, the context, coordination, governance, and disbursal administration surrounding them have not kept pace.
For many developers, agents are already delivering applicable value. At the organizational level, the economics are much harder to prove. The expenses exterior elsewhere – in review, rework, security, infrastructure, and spend.
Which agents can admission business code? Where can data go? Which output requires individual review? What happened during an delegate was operating remotely? Who approved the resulting change, and how was it verified?
Fragmentation at this flat isn’t fair irritating. It makes application betterment harder to understand, measure, and govern at exactly the item whenever additional of the activity is being delegated.
The bottleneck is shifting alongside the work
Code that’s evidently incorrect gets caught quickly. That part of the scheme motionless works. The harder issue is code that’s nearly right: plausible, capable of passing a superficial check, but quietly carrying a bad assumption or architectural inconsistency that won’t exterior until it’s expensive.
As agents obtain on additional of the execution, the bottleneck shifts from producing alter to understanding, verifying, and owning it. Code becomes cheaper to create but additional costly to verify. Agent action becomes easier to commencement but harder to coordinate, audit, and explain.
And during the activity can be delegated, accountability cannot. An delegate volition not get the call at 3:00 am whenever item breaks. The duty for what ships motionless belongs to the group and organizations that container it.
This is why authority becomes harder, not easier, as AI improves. A additional capable example may create improved output. It does not established organizational policy, maintain provenance, provision disbursal visibility, or decide who accepts duty for the resulting change.
The forthcoming is multi-vendor
Multi-vendor assistance is a foundational scheme regulation of JetBrains Air, shaping how the scheme is being built from the outset.
We don’t accept this market volition consolidate any period soon. Models change in what they’re fine at, and rankings alter all few months. Teams inner the identical business already create distinct choices, and they are frequently correct to do so. Standardizing on one AI vendor today method making a multi-year dedication in a market that won’t appearance the identical next quarter.
Keeping the options open is the sensible item to do. The issue is what openness currently costs. Every new model, agent, or assistance an institution adds takes distant a small additional visibility into its own betterment work. Context doesn’t transport complete between tools. Spend can’t be attributed. Policies have to be rebuilt for all service.
Organizations should not have to choose between using the finest accessible tools and understanding what is happening inner their own engineering. That trade-off exists since nothing in the current stack was built to sit complete multiple vendors at once.
This is the activity JetBrains has taken on. We build our own agent, and we average to create it excellent. But JetBrains Air does not necessitate customers to use ours, and our scheme does not depend on which example provider leads the rankings this quarter. We have no logic to create the ecosystem smaller than it is.
What we can recommendation alternatively is one location to run, see, govern, and document for agentic betterment throughout all model, agent, and assistance – for the developer, the team, and the organization.
Supporting multiple models and agents is the floor, not the ceiling. The part that matters is what sits complete them: shared context, one set of policies, a sole disbursal view, and a document of what happened, despite of which vendor produced the change.
Why JetBrains?
Multi-vendor choice solves lone part of the problem. Agents additionally need dependable application intelligence.
JetBrains brings 26 years of engineering intellect to the problem, assisting developers comprehend the construction and behavior of complex software, not merely create additional of it. That deterministic code intellect provides a basis for making agentic activity additional reliable, efficient, and understandable throughout distinct models and agents. We are seeing promising results from giving AI agents admission to deterministic code intelligence.
This is an financial advantage as fine as a specialized one. Agents expend period and prosperity rediscovering data the codebase already contains. An delegate that can recover that cognition is cheaper and additional exact than one that has to rebuild it. Because intellect does not pertain to one model, the advantage can broaden throughout supported agents and services.
We are additionally going through the identical passage as the organizations we build for, adopting agents internally, redesigning workflows, and learning anywhere idiosyncratic efficiency gains translate into improved application shipment and anywhere they merely move activity elsewhere.
What comes next
JetBrains Air volition create through a rolling sequence of releases. We volition be definitive concerning what customers can use now, what is entering preview, and what remains part of our longer-term direction.
Over time, JetBrains Air volition broaden additional into mobile and distant experiences, allowing group to initiate, monitor, review, and continue agentic activity as it moves between environments. The goal is not to reproduce the IDE on all surface. We are making the correct environment and controls accessible anywhere decisions need to be made.
We volition additionally bring JetBrains’ intellect into additional agentic workflows. This includes richer environment drawn from code, architecture, repositories, runtime behavior, and organizational knowledge, as fine as improved ways to path activity between developers, models, agents, and services.
More activity volition be triggered by repository events, schedules, and shipment processes fairly than by a developer beginning an publishing company and issuing a prompt. JetBrains Air volition provision the intelligence, oversight, and individual authority these workflows necessitate throughout surfaces and services.
We volition not name forthcoming products before their range and preparedness are prepared to be confirmed. With all release, we volition explain what works, how it connects, and what’s motionless in progress.
Where JetBrains Air is going
The companies that win in adopting AI volition not necessarily be those that create the most code or deploy the most agents. They volition be those that can develop experimentation without losing quality, context, disbursal discipline, or individual understanding.
JetBrains Air is our dedication to construction for that reality. It expands JetBrains from the developer workbench into a scheme of products connecting developers, agents, teams, and organizations.
The goal is not additional code. It is application that developers, teams, and organizations can understand, verify, and remain behind.
