Senior GenAI Safety Scientist

A*STAR - Agency for Science, Technology and Research; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

A*STAR - Agency for Science, Technology and Research's Trustworthy AI team at CFAR, IAIC, seeks motivated scientists to research the safety of generative AI and multi-agent systems. Working with regulators and clinicians, the team develops evidence, testing tools and guidance for real-world deployment.

GenAI and multi-agent systems, such as clinical chatbots built on LLMs and VLMs interpreting medical images, pose risks not covered by current regulations.

Qualifications

• PhD level candidate with strong research track record.
• Proficient in Python and PyTorch with hands-on DL experience.
• Experience in AI safety, adversarial ML or trustworthy ML is desirable.
• Excellent communication and ability to work with clinicians, regulators and engineers.

Responsibilities

• Design and conduct systematic red-teaming and risk assessments of GenAI and multi-agent systems.
• Develop guardrails and risk-mitigation methods, including detection and flagging of high-risk outputs.
• Collaborate with regulatory and clinical partners to translate findings into regulatory-grade tools and guidance.

Skills

Python

PyTorch

AI safety

Adversarial ML

Trustworthy ML

LLMs

VLMs

Multi-agent systems

Communication

Teamwork

Education

PhD in Computer Science / Electrical & Electronic Engineering / Computer Engineering / Biomedical Engineering or related field

Tools

PyTorch

Job description

A*STAR - Agency for Science, Technology and Research's Trustworthy AI team at CFAR, IAIC, seeks motivated scientists to research the safety of generative AI and multi-agent systems. Working with regulators and clinicians, the team develops evidence, testing tools and guidance for real-world deployment.

GenAI and multi-agent systems, such as clinical chatbots built on LLMs and VLMs interpreting medical images, pose risks not covered by current regulations.