Machine learning for Software Engineering
Machine learning and AI have recently revolutionized research and practice in areas such as Image Processing, Natural Language Processing and autonomous systems. Recently, machine learning techniques, mostly stemming from the NLP domain, have been applied on software engineering problems, with promising results. Despite this, we are just at the beginning: the field is ripe with opportunity for high impact and exiting applications!
The research will take place within the context of the FASTEN project, a large scale project aiming to apply static analysis techniques at the software ecosystem level. Example projects may include (but are not limited to):
- Analyses to identify security, licensing and other problems in dependency networks
- Machine-augmented static analyses
- Machine representations of software dependency networks
- Graph Neural Networks for modeling source code
If you apply for a PhD position:
- A Master’s degree (or equivalent) in computer science
- A healthy dose of creativity and ability to come up with fresh ideas
- A strong interest in Machine Learning. Demonstratable experience is a big plus.
- Strong programming and software engineering skills
- Good communication and presentation skills in English
- Ability to work in an international team
In addition, if you apply for a postdoc position:
- A PhD degree in Software Engineering or in Machine Learning and related fields
- Strong publication record in SE / ML or related conferences
How to apply?
Please send your application material, including a CV, motivation letter and any material you think will be supporting your application (e.g. Github repository, references etc), to Georgios Gousios (email@example.com).
Please contact Georgios directly for any question recarding the positions.
The research environment
The Software Analytics Lab at TU Delft, led by Georgios Gousios, conducts research at the intersection of software engineering, big data and machine learning. Focus areas include:
- Static analysis and its applications
- Real-time software analytics
- Large scale developer collaboration
Notable projects that came out of the group include GHTorrent, TravisTorrent, CodeFeedr and FASTEN. The lab has strong links to the industry, working together on projects with companies such as Microsoft, Facebook and ING. The Software Analytics lab comprises a young, ambitious, and international research group. The working language is English (no Dutch is necessary).
The Software Engineering Research Group (SERG), led by Arie van Deursen, is part of the department of Software Technology, faculty of Electrical Engineering, Mathematics, and Computer Science, and a member of the research school IPA. The group employs approximately 30 people, comprising full, associate, and assistant professors, lecturers, postdocs, and PhD students. The main research areas for the group include but are not limited to software analytics, software testing, software quality and maintenance, software evolution, search-based software engineering and machine learning for software engineering. The involved researchers have developed numerous techniques to make it easier for software developers to (1) understand, (2) maintain and (3) test existing software systems. More information about the group can be found here: https://se.ewi.tudelft.nl
Located in a charming college town, TU Delft is the largest and oldest public technological university in the Netherlands. The university is regularly ranked among the most highly-rated worldwide for engineering and technology. Information about academic careers in the Netherlands and working at TU Delft can be found at www.factcards.nl and www.tudelft.nl/en/about-tu-delft/working-at-tu-delft/coming-to-the-netherlands-tu-delft/.