Deskripsi Pekerjaan
Are you passionate about the ethical implications of Artificial Intelligence? Nanyang Technological University (NTU) is seeking a dedicated AI Engineer (Safety) to lead the charge in securing AI models for the Southeast Asian region. In this pivotal role, you will work at the intersection of technology and policy to ensure that AI systems are robust, fair, and aligned with human values. You will collaborate with world-class researchers to develop safety protocols, conduct rigorous adversarial testing, and mitigate risks associated with Large Language Models and generative AI. This is a unique opportunity to shape the future of AI governance while contributing to cutting-edge research at one of Asia's leading universities.
As a member of our team, you will play a critical role in defining the safety standards for AI applications in our region, ensuring that innovation does not come at the cost of safety and security. We are looking for a proactive problem solver who is committed to building trustworthy AI systems that benefit society.
Tanggung Jawab
- Design and implement safety guardrails for AI models to prevent misuse and bias.
- Conduct red-teaming and adversarial testing on generative AI systems to identify vulnerabilities.
- Analyze model outputs to detect and mitigate toxic or harmful content.
- Collaborate with interdisciplinary teams to translate technical findings into policy recommendations.
- Develop and maintain documentation for safety standards and compliance frameworks.
- Research and evaluate emerging safety methodologies and alignment techniques.
Kualifikasi
- Master’s or PhD degree in Computer Science, Engineering, or a related field.
- Strong proficiency in Python and deep learning frameworks (PyTorch, TensorFlow).
- Deep understanding of Natural Language Processing (NLP) and Large Language Models (LLMs).
- Experience with AI alignment concepts and safety best practices.
- Excellent written and verbal communication skills in English.
- Ability to think critically about system design and edge-case scenarios.