About the Role

We are looking for a Researcher in Trustworthy AI to conduct research and development on making Artificial Intelligence systems secure, robust, privacy-preserving, explainable, transparent, and accountable.

The role will focus on developing novel methods, algorithms, and evaluation frameworks to assess and improve the trustworthiness of AI systems throughout their lifecycle. The successful candidate will work across areas such as AI security and robustness, privacy-preserving AI, Explainable AI (XAI), AI assurance and auditing, fairness, and Responsible AI, including emerging trustworthiness challenges associated with Generative AI, LLMs, and agentic AI systems.

The researcher will also have the opportunity to contribute to SHIELD-6G, an EU-funded collaborative R&D project, where our team is actively involved in research on Trustworthy AI, continuous AI auditing and assurance, AI security, explainability, and compliance of AI-enabled systems. The role will involve collaboration with international research and industry partners and the validation of research outcomes in next-generation communication and 6G environments.

Key Responsibilities

• Conduct research on Trustworthy AI, with a focus on security, privacy, robustness, explainability, transparency, fairness, and accountability.

• Design and develop methodologies and technical frameworks for assessing and continuously monitoring the trustworthiness of AI/ML systems.

• Research AI security and robustness, including adversarial attacks, data poisoning, model manipulation, privacy attacks, distribution shifts, and corresponding mitigation techniques.

• Research and develop privacy-preserving AI/ML approaches, including federated learning, differential privacy, and secure collaborative learning.

• Develop and evaluate Explainable AI (XAI) methods to improve the interpretability and transparency of AI systems, including assessment of explanation quality, reliability, and faithfulness.

• Develop methodologies for AI assurance and continuous auditing, including trustworthiness metrics, risk indicators, technical controls, and automated assessment mechanisms.

• Investigate fairness, bias, and Responsible AI methodologies and their integration into AI development and evaluation processes.

• Research trustworthiness challenges associated with Generative AI, Large Language Models (LLMs), RAG systems, and autonomous/agentic AI systems.

• Develop mechanisms supporting model and data provenance, traceability, integrity, and auditability throughout the AI lifecycle.

• Translate relevant AI governance, security, and regulatory requirements into measurable technical requirements and evaluation criteria.

• Design experiments, benchmarks, and PoCs to validate developed Trustworthy AI methods in realistic environments.

• Contribute to scientific publications, patents, international R&D projects, and relevant Trustworthy AI standardization activities.

Required Qualifications

• MSc or PhD, or equivalent research experience, in Artificial Intelligence, Computer Science, Electrical/Electronics Engineering, Cybersecurity, Data Science, or a related field.

• Strong understanding of machine learning, deep learning, and modern AI systems.

• Research or hands-on experience in at least one Trustworthy AI area such as:

  • Explainable AI (XAI)

  • Privacy-preserving machine learning

  • AI/ML security and adversarial machine learning

  • Robust machine learning

  • Fairness and bias mitigation

  • AI assurance, auditing, or model monitoring

  • Responsible AI

• Strong programming skills, preferably in Python, and experience with AI/ML frameworks such as PyTorch, TensorFlow, or equivalent.

• Ability to design experiments, define quantitative evaluation metrics, and critically assess AI models beyond conventional performance metrics.

• Ability to follow scientific literature and translate research ideas into algorithms, prototypes, and experimental studies.

• Familiarity with Generative AI and LLM-based systems and their emerging trustworthiness challenges.

• Knowledge of AI governance and assurance frameworks such as the EU AI Act, NIST AI RMF, or ISO/IEC 42001 is considered an advantage.

• Experience with scientific publications, international R&D projects, patents, or standardization activities is considered an advantage.

• Strong analytical, problem-solving, and research skills, with good written and spoken English.

​

​

​

​

​

​