PhD Position Validating Automatically Generated Code

AI-powered coding tools are set to revolutionize how we develop software. These AI tools are based on machine learning (ML) models trained on massive collections of open-source code and allow developers to easily (re)use the programming knowledge aggregated by these ML models.
Although LLMs can generate code, recent studies have shown they have several important limitations and problems with the generated code. For instance, Copilot, which is based on Codex, has been found to generate code snippets with small bugs that are challenging to fix manually. Additionally, LLM models may produce significantly different and even unexpected results when queried with semantically equivalent yet differently phrased requests.
All these points have the same underlying question that needs to be answered for the AI tools to become useful: How do we validate the generated code?
The main focus of this PhD position is to answer this question.
Note: This PhD position is part of the AI for Software Engineering lab (AI4SE), a collaboration between JetBrains and Delft University of Technology. Hence, the prospective PhD student will also work closely with researchers and developers of JetBrains. More information is available on https://lp.jetbrains.com/research/ai-for-se/.

Requirements:

Needs-to-have:

• A Master's Degree (or equivalent) in Computer Science, with a focus on Software Engineering, Software Testing, Machine Learning, or similar areas.
• Proficiency in JVM-based languages such as Java, Kotlin, or Scala, ideally supported by contributions to open-source projects.
• Effective communication skills in English, both written and spoken.
• A team-centric attitude, able to work in international collaborations with researchers and developers.
• An independent, proactive approach combined with pragmatic practical creativity.

Nice-to-have:

• Knowledge of Large Language Models in Software Engineering or experience in their practical use.
• Experience with static and dynamic program analysis techniques, such as type systems or symbolic execution.
• A readiness to take the lead on projects.
• An inclination to mentor and guide junior colleagues.

Doing a PhD at TU Delft requires English proficiency at a certain level to ensure that the candidate is able to communicate and interact well, participate in English-taught Doctoral Education courses, and write scientific articles and a final thesis. For more details please check the Graduate Schools Admission Requirements.

Salary Benefits:

Doctoral candidates will be offered a 4-year period of employment in principle, but in the form of 2 employment contracts. An initial 1,5 year contract with an official go/no go progress assessment within 15 months. Followed by an additional contract for the remaining 2,5 years assuming everything goes well and performance requirements are met.
Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities, increasing from € 2770 per month in the first year to € 3539 in the fourth year. As a PhD candidate you will be enrolled in the TU Delft Graduate School. The TU Delft Graduate School provides an inspiring research environment with an excellent team of supervisors, academic staff and a mentor. The Doctoral Education Programme is aimed at developing your transferable, discipline-related and research skills.
The TU Delft offers a customisable compensation package, discounts on health insurance, and a monthly work costs contribution. Flexible work schedules can be arranged. 
For international applicants, TU Delft has the Coming to Delft Service. This service provides information for new international employees to help you prepare the relocation and to settle in the Netherlands. The Coming to Delft Service offers a Dual Career Programme for partners and they organise events to expand your (social) network.

36 - 40 hours per week

Mekelweg 2