The ASGAARD lab develops AI-powered techniques for detecting and understanding software defects, with a focus on quality assurance (QA) in interactive systems such as video games, web applications, and AI-driven software.
🎮 Intelligent Quality Assurance for Video Games
2021–present
Modern video games are visually complex and difficult to test automatically. We investigate how AI can help detect gameplay glitches and visual defects at scale.
- Detect gameplay glitches using Vision-Language Models (VLMs) and Large Multimodal Models (LMMs)
- Develop automated techniques for visual regression testing and gameplay anomaly detection
- Analyze gameplay videos to automatically generate bug reports
Key artifacts
VideoGameQA-Bench
Industry collaborators
EA, Prodigy Education, Sony Interactive Entertainment, Ubisoft La Forge
🌐 Intelligent Quality Assurance for Web Applications and Browsers
2021–present
Modern web applications rely on complex rendering engines and dynamic front-end frameworks.
- Detect cross-browser rendering inconsistencies
- Identify GUI layout defects and visual regressions
- Analyze screenshots across browsers and devices
Key artifacts
HTML5 canvas bug taxonomy
Industry collaborators
🤖 Software Engineering for AI & Foundation Models
2023–present
This research investigates the intersection of software engineering and foundation models.
- FM4SE: using foundation models to support software engineering tasks
- SE4FM: applying engineering practices to the development and maintenance of AI systems
- Study practitioner perspectives and development practices around AI systems
Key artifacts
Practitioner studies
🎥 Mining Gameplay Videos for Bug Discovery
2018–2022
Gameplay videos shared online contain valuable evidence of real-world software bugs.
- Automatically identify videos that demonstrate gameplay glitches
- Mine large video repositories such as YouTube for bug discovery
- Study how gameplay videos can support debugging and QA processes
Key artifacts
Video-based bug discovery techniques
⚡ Empirical Performance Engineering & Microbenchmarking
2015–2025
This research investigates how to detect and diagnose performance regressions in large software systems.
- Detect performance regressions using system metrics and logs
- Analyze architectural signals to identify performance problems early
- Study microbenchmarking practices and performance testing strategies
Key artifacts
Benchmarking studies
💬 Improving Post-Release Support
2017–2023
Post-release support plays a critical role in maintaining user satisfaction and software quality.
- Study how developers respond to user reviews in app stores
- Analyze how developer responses affect user ratings
- Improve bug reporting and triaging processes
Key artifacts
Bug report analysis techniques
🚀 Studying the Impact of Release Strategies
2015–2019
Modern development practices such as DevOps enable rapid and frequent software releases.
- Study how release strategies affect user-perceived software quality
- Analyze urgent updates and zero-day patches in video games
- Investigate why some mobile app releases receive negative user feedback
Key artifacts
User perception analyses