> Markdown version of [/jobs/ext/1369920-phd-in-resilient-machine-learning-and-formal-methods](https://www.wearedevelopers.com/jobs/ext/1369920-phd-in-resilient-machine-learning-and-formal-methods). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # PhD in Resilient Machine Learning and Formal Methods - **Company:** Eindhoven University of Technology - **Location:** Eindhoven, Netherlands (Remote available) - **Contract:** Temporary contract - **Skills:** Artificial Intelligence, Computer Programming, Data Distribution Service, Systems Analysis, Python (Programming Language), Machine Learning, Tensorflow, Reinforcement Learning, Pytorch, Information Technology, Formal Methods - **Published:** July 22, 2026 - **Apply:** https://nl.indeed.com/viewjob?jk=e4beacb7db3aa1ad ## About the Role Are you passionate about the foundations of safe and resilient AI? Do you want to develop machine learning approaches that not only withstand adverse conditions but actively learn from their own failures - and can you back those systems with rigorous formal guarantees? We are looking for a motivated and talented PhD candidate to join a unique interdisciplinary project at the intersection of machine learning and formal methods., * MSc degree (or near completion) in computer science, mathematics, artificial intelligence, or a closely related discipline. * A strong foundation in machine learning, including familiarity with statistical learning theory and probabilistic models; prior exposure to notions of robustness, resilience, or uncertainty quantification is an advantage. * Mathematical maturity and experience with formal or rigorous reasoning; prior knowledge of formal methods, logics, or verification is beneficial but not required - genuine curiosity and willingness to develop these skills is essential. * Good programming skills preferably in Rust or Python; practical experience with Python-based machine learning frameworks (e.g. PyTorch) is preferred. * A motivated, creative, and self-driven working style with the ability to collaborate effectively in an interdisciplinary team. * Motivated to contribute to teaching and to coach students as part of your professional development. * Fluent in spoken and written English (at least C1 level). ## Description Machine learning models deployed in real-world environments inevitably face noisy data, distribution shifts, and situations their training never anticipated. Existing machine learning research focuses on limiting the impact of such disturbances, rendering such models robust. The PhD project aims to go the next step, closing the loop between failure and adaptation by developing resilient machine learning. The developed methods will detect when something has gone wrong, learn from failures and mispredictions, and recover to a stable and reliable operating state. Formal methods ensure that learning and recovery are performed with a provable quality of service. The project is conducted in collaboration with two clusters at the Department of Mathematics and Computer Science of TU/e: Data and Artificial intelligence. Novel learning algorithms will be developed, capable of operating in reactive, online settings where the data distribution may shift over time. Key challenges include detecting prediction failures, designing self-correcting update mechanisms that exploit past errors, and evaluating resilience empirically on challenging benchmarks. The project team has access to the national computing infrustracture, and TU/e HPC cluster SPIKE-1. Formal system analysis. Rigorous formal methods will ensure that learnt machine learning methods and the resulting systems are indeed resilient, aiming at improved dependability and trustworthiness. Formal notions of resilience have to be developed along with algorithms to check resilience of machine learning models. Research is conducted in the fields of automated reasoning, probabilistic verification, and reinforcement learning to derive provable guarantees on resilience. You will work at the interface of these two highly timely perspectives, contributing to both the algorithmic development and the formal analysis. The project involves regular collaboration among all three supervisors and will result in publications at top venues in machine learning, artificial intelligence, and formal methods. The position is embedded in the vibrant research environment of the newly established Center for Safe AI. PhD candidate will be formally employed with the Data and AI cluster at the Department of Mathematics and Computer Science and supervised by Mykola Pechenizkiy, Cassio de Campos and Clemens Dubslaff., A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you, * Official grade transcripts of your BSc and MSc education, listing all courses taken and grades obtained. * Curriculum vitae, including a list of your projects, publications, and any other relevant items * A copy of or a link to your MSc thesis or an example of your academic writing if the MSc thesis cannot be shared. Ensure that you submit all the requested application documents. We give priority to complete applications. We look forward to receiving your application and will screen it as soon as possible. The vacancy will remain open until the position is filled. ## Related Videos - [Staying Safe in the AI Future](https://www.wearedevelopers.com/videos/521-staying-safe-in-the-ai-future) - [Photonic Computing: Programming a New Class of AI Accelerators (incl. Live Coding)](https://www.wearedevelopers.com/videos/100196-photonic-computing-programming-a-new-class-of-ai-accelerators-incl-live-coding) - [Machine learning in the browser with TensorFlowjs](https://www.wearedevelopers.com/videos/155-machine-learning-in-the-browser-with-tensorflowjs) - [When testing just doesn’t cut it](https://www.wearedevelopers.com/videos/720-when-testing-just-doesn-t-cut-it) - [AI Meets Hoare Logic: Revolutionizing Software Testing with Formal Methods](https://www.wearedevelopers.com/videos/1644-ai-meets-hoare-logic-revolutionizing-software-testing-with-formal-methods) - [Getting Started with Machine Learning](https://www.wearedevelopers.com/videos/260-getting-started-with-machine-learning) ## Related Articles - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [MLOps – What’s the deal behind it?](https://www.wearedevelopers.com/magazine/125-mlops-what-s-the-deal-behind-it) - [MLOps And AI Driven Development](https://www.wearedevelopers.com/magazine/82-mlops-and-ai-driven-development) - [Building AI Solutions with Rust and Docker](https://www.wearedevelopers.com/magazine/494-building-ai-solutions-with-rust-and-docker) - [MLops – Deploying, Maintaining And Evolving Machine Learning Models in Production](https://www.wearedevelopers.com/magazine/115-mlops-deploying-maintaining-and-evolving-machine-learning-models-in-production) - [Dev Digest 121 - AI goes offline](https://www.wearedevelopers.com/magazine/456-dev-digest-121-ai-goes-offline)