My Research Vision

Software is eating the world but more software means more complexity. In the current decade, software no longer operate in isolation, but interact with other one and execute in increasingly intricate ecosystem. As a result, there are more and more onerous burden of developing and maintaining software systems on software developers. To reduce the increasingly burden, we need better ways to support developers on their daily tasks.

Towards this goal, my research aim to leverage AI and data mining to tackle real-world challenges in software engineering and pushing the boundaries of programming technologies. My goal to is to build trustworthy AI-powered tools for supporting developers on software engineering tasks (AI4SE). Particular, I'm trying to explore the following research questions:

  • (1) How reliable and depandable AI4SE tools?
  • (2) How can we improve their reliability and trustworthiness?
  • (3) How can we improve developers' trust on these tools?

    • More specifically, I mainly work on bugs/vulnerability detection, management and fixing, which cost ~50% of the programming time of a software developer on average. My current focus (for my PhD) is on improving the trustworthiness of systems. I have also previously worked on securing by effectively managing vulnerabities from third-party libraries. Besides, I am also interested in investiagating and advancing the trustworthiness/reliablity of .

    Program Repair

    Program Repair, a.k.a Automated Bug Fixing is an emerging technology to alleviate the onerous burden of manually fixing bugs on developers. A substantial number of APR techniques have been proposed over the years with several breakthroughs that inspired potential practical adoption of APR. Unfortunately, developer's trust on APR-generated patches is still a challenge for achieving practical adoption of APR on industry. This theme aim to investigate unknown issues regarding the trustworthiness of APR systems for enhancing the developers' trust of such systems. If you are interested in this theme, see more about my research here.

    Software Security

    Modern software development involves the heavy use of APIs and third-party components. The reliance increase security risks of software system as the components can contain exploitable vulnerabilities. This theme aim to mitigate these risks by creating an advanced software composition analysis for scanning dependency hierarchies and building AI models to analyse code and document repository data and flag vulnerabilities. If you are interested in this theme, see more about my research here.

    Large Language Models for Software Engineering

    Large Language Models (LLMs) have demonstrated remarkable performance on understanding and generating human-like responses, rendering them highly suitable for various applications such as code summarization or code generation. Unfortunately, while LLMs have shown great promise, the reliability of LLM-generated responses are still questionable. This theme is to explore unknown issues regarding the reliability of LLM-generated responses and propose effective solutions for identified issues. If you are interested in this theme, see more about my research here.

    Communication and Collaboration

    Our group and I loves to collaborate and communicate. If you want to know more about our works or share awesome ideas, please feel free to contact me via email congthanh.le student.unimelb.edu.au