JetBrains releases Koog: open source framework for AI agents
Koog from JetBrains, with modular architecture for flexible AI agent development, enables precise control over workflows and agent activities.
(Image: Elnur/Shutterstock.com)
JetBrains has released the Koog framework, an open-source software for developing AI agents in the JVM ecosystem using Kotlin. Presented at KotlinConf, Koog provides tools to develop autonomous agents.
Modular architecture
According to the blog post, a key feature of Koog is its modular architecture, which allows developers to extend agent functions through a system of composable modules. This modularity should ensure that specific functions such as the streaming of Large Language Models (LLMs) or the generation of structured output can be seamlessly integrated and adapted. Developers can create custom tools that give agents access to external systems and resources, enabling interaction with different data sources or services.
Koog supports workflow customization by allowing developers to flexibly adjust the order and logic of the tasks an agent handles. According to the blog post, this is possible with the help of configurable workflows that can be tailored to the specific requirements and goals of the application.
Videos by heise
Agents under control
Koog also offers comprehensive monitoring and tracking functions that enable detailed observation of agent activities, LLM calls and the use of tools.
Koog is available as an open-source project on GitHub and is designed to enable developers to create agents that can handle complex tasks automatically.
(Image: TechSolution/Shutterstock)
Following the great success of the first betterCode() GenAI, the online conference on AI-supported software development will take place again on June 26.
The organizers iX and dpunkt.verlag have updated the conference program and further improved it based on feedback. It offers the following presentations:
- Software development with Copilot, ChatGPT and Co
- What's new in AI coding tools?
- Testing software with AI support
- Defeating dinosaurs with ChatGPT – LLMs for analyzing old systems
- Strengths and weaknesses of AI-supported, secure software development
- Legal aspects of AI-supported software development
(mdo)
