JetBrains and GitHub Integrate AI-Powered Coding Agents into Popular IDEs
In a significant advancement for integrated development environments (IDEs), JetBrains and GitHub have unveiled AI coding agents designed to streamline and enhance the software development process. These tools, now integrated into widely used IDEs, represent a shift towards more autonomous and intelligent coding assistance.
JetBrains has officially launched “Junie,” its AI coding agent, across several of its IDEs, including IntelliJ IDEA, PyCharm, WebStorm, PhpStorm, GoLand, and RubyMine. This move follows GitHub’s recent general availability release of its Copilot agent mode for Visual Studio Code (VS Code) on April 4, 2025. Both companies aim to provide developers with tools that can handle complex coding tasks through natural language prompts, moving beyond the capabilities of traditional code completion assistants.
Unlike earlier AI assistants that offered inline code suggestions, these new agents can execute entire tasks based on user instructions. For instance, a developer can prompt the agent to “create a web application in Python,” and the agent will generate the necessary code, manage dependencies, and even run tests. Diego Lo Giudice, an analyst at Forrester Research, describes this as an early form of “agentic AI,” where the system can autonomously perform multi-step workflows. However, he notes that the broader infrastructure to support fully autonomous AI agents is still under development, with initiatives like the Model Context Protocol aiming to address this gap.
JetBrains has integrated Junie into its IDEs under a unified subscription model, which includes a free tier offering unlimited code completion and access to local AI models. This tier also provides credit-based usage of cloud-based AI assistance. Higher-tier subscriptions, such as AI Pro and AI Ultimate, offer increased usage quotas and additional features. This approach mirrors GitHub’s strategy, which also includes AI agents in its existing subscription plans.
The introduction of these AI agents is not without challenges. Kyler Middleton, a principal software engineer at healthcare technology company Veradigm, shared her experience with GitHub’s Copilot agent mode. She tasked the agent with refactoring existing code, and while it attempted to fulfill the request, it introduced new errors and deviated significantly from the original code structure. Middleton likened the experience to working with a junior engineer who requires supervision and feedback, emphasizing that these agents still need human oversight to ensure code quality and functionality.
JetBrains’ Junie aims to address some of these concerns by offering a more collaborative experience. The agent can propose plans, adhere to development guidelines, and perform complex tasks with greater control, allowing developers to maintain a human-in-the-loop approach. Junie is currently compatible with IntelliJ IDEA Ultimate, PyCharm Pro, WebStorm, and GoLand, with plans to extend support to PhpStorm, RustRover, and RubyMine in the near future.
In addition to Junie, JetBrains has updated its AI Assistant, enhancing its capabilities to support developers throughout the entire development process. The latest version includes support for new models like Anthropic’s Claude 3.7 Sonnet and Google’s Gemini 2.5 Pro, improved code completion, greater context awareness, and a new edit mode that enables multi-file edits directly from the chat interface. These enhancements aim to accelerate coding workflows and reduce repetitive tasks.
The integration of AI agents into IDEs marks a significant step in the evolution of software development tools. By enabling developers to delegate complex tasks to intelligent agents, these tools have the potential to increase productivity and allow developers to focus on higher-level design and problem-solving. However, as the technology continues to mature, the importance of human oversight and collaboration remains paramount to ensure the quality and reliability of the generated code.
As AI continues to permeate various aspects of software development, the collaboration between human developers and AI agents is poised to redefine traditional workflows, offering new opportunities for innovation and efficiency in the coding process.
