Introduction to AI4SE
The collaboration between JetBrains and Delft University of Technology (TU Delft) has given birth to AI for Software Engineering (AI4SE), a research partnership that aims to harness the power of Artificial Intelligence (AI) to transform the software engineering landscape. In this article, we will delve into the achievements of AI4SE in 2025, exploring the five dedicated research tracks that are driving innovation in the field.
AI4SE Overview
Launched in late 2023, AI4SE brings together leading labs, industry-grade tools, and five research tracks to turn advances in AI into practical gains for developers and learners worldwide. The research partnership is a significant step forward in the evolution of software engineering, with the potential to revolutionize the way software is built, tested, and taught.
Research Tracks
The five research tracks of AI4SE are designed to explore the applications of AI in software engineering, with a focus on autonomous AI agents, large language models, interactive IDEs, runtime information, and intelligent teaching assistants. These tracks are:
- Testing and Evaluating LLMs and SWE Agents
- Large Language Model Adaptation for Coding Tasks
- Interactive and Aligned IDEs in the LLM Era
- Utilizing Runtime Information to Improve Development Processes
- Intelligent Teaching Assistant in Programming Education
AI4SE in 2025: Highlights
In 2025, AI4SE made significant progress in each of the five research tracks. The Testing and Evaluating LLMs and SWE Agents track explored the use of autonomous AI agents in coding, testing, and automating workflows. The Large Language Model Adaptation for Coding Tasks track focused on adapting and personalizing LLMs for different IDE users’ coding tasks. The Interactive and Aligned IDEs in the LLM Era track aimed to build simple, non-disruptive tools that bring AI-powered code generation and explanation directly into developers’ existing IDE workflows.
Conclusion
In conclusion, AI4SE is revolutionizing the software engineering landscape by harnessing the power of AI to transform the way software is built, tested, and taught. With its five dedicated research tracks, AI4SE is driving innovation in the field, with significant progress made in 2025. As we look to the future, it is clear that AI4SE will continue to play a major role in shaping the software engineering industry.
Frequently Asked Questions
- What is AI4SE? AI4SE is a research partnership between JetBrains and Delft University of Technology (TU Delft) that aims to harness the power of Artificial Intelligence (AI) to transform the software engineering landscape.
- What are the five research tracks of AI4SE? The five research tracks of AI4SE are Testing and Evaluating LLMs and SWE Agents, Large Language Model Adaptation for Coding Tasks, Interactive and Aligned IDEs in the LLM Era, Utilizing Runtime Information to Improve Development Processes, and Intelligent Teaching Assistant in Programming Education.
- What is the goal of the Large Language Model Adaptation for Coding Tasks track? The goal of the Large Language Model Adaptation for Coding Tasks track is to adapt and personalize LLMs for different IDE users’ coding tasks, evaluating the LLMs’ emerging capabilities and ensuring that the outputs are timely, safe, and relevant.
- How does AI4SE plan to improve development processes? AI4SE plans to improve development processes by utilizing runtime information to assist in AI engineering and AI observability.
- What is the role of the Intelligent Teaching Assistant in Programming Education track? The Intelligent Teaching Assistant in Programming Education track aims to build a smart AI teaching assistant that gives students personalized, context-aware help with programming.
