Data Scientist, Responsible AI

About GovTech

The Government Technology Agency (GovTech) is the lead agency driving Singapore’s Smart Nation initiatives and public sector digital transformation. As the Centre of Excellence for Infocomm Technology and Smart Systems (ICT & SS), GovTech develops the Singapore Government’s capabilities in Data Science & Artificial Intelligence, Application Development, Smart City Technology, Digital Infrastructure, and Cybersecurity. GovTech aims to transform the delivery of Government digital services by taking an "outside-in" view, putting citizens and businesses at the heart of everything we do. GovTech offers a purposeful career to make lives better, empowering its people to master their craft through continuous learning and development. Join GovTech to play a part in Singapore’s vision to build a Smart Nation and embark on a meaningful journey to build tech for public good.

Who We Are

The Responsible AI team, part of GovTech's Artificial Intelligence Tech Practice, seeks to develop technical capabilities to ensure AI systems are safe, trustworthy, and ethical. We aim to be at the forefront of the AI safety and responsible AI space through learning, experimentation, and iteration.

Our specific focus areas are:

• Conducting safety testing of LLM/MLM systems and fairness assessments of high-impact AI systems.
• Developing safety guardrails for government LLM/MLM systems.
• Improving understanding of responsible AI across the public sector.

We work closely with AI product teams to enable them to meet AI safety and responsible AI standards with ease.

What You Will Be Working On

As a Data Scientist in the Responsible AI team, you will:

• Focus on reading up on the latest developments in the AI safety and/or algorithmic fairness space.
• Experiment with new open-source tools or methods to raise the level of AI safety and fairness in the government’s AI systems.
• Conduct fairness assessments of high-impact government AI systems and propose how best to mitigate bias if any.
• Have the opportunity to finetune new models for safety alignment to support internal safety needs.
• Work closely with AI product teams and other public officers to help them understand responsible AI at a deeper level and to help them adopt safety testing and guardrails into their own products.
• Collaborate with technical partners, including the private sector, academic institutions, and international players.
• Present your work and share expertise with the wider community on LLM safety testing, guardrails, and fairness assessments.

Who We Are Looking For

• Demonstrated experience with end-to-end data science projects, from scoping, development, deployment, and monitoring.
• Strong technical understanding of LLMs, especially how they work, how to do prompt engineering effectively, and how to build apps with them.
• Good understanding of key AI safety and/or algorithmic fairness concepts, especially in how to operationalise and apply them to AI systems.
• Competent in Python programming, with some familiarity of cloud deployments for ML models and LLMs.
• Strong interest in AI safety or responsible AI, especially given the need to read up independently on these areas and continuously explore and experiment.
• Self-motivated, thinks proactively and creatively about how to solve challenges.
• Desire to build tech for the public good.

Strong preference for applicants with informal or formal experience with any of the work below:

• Conducting LLM safety testing, including jailbreaking or prompt injections.
• Developing guardrails for LLMs or using equivalent tools (e.g., NeMo, LLM-Guard).
• Developing fairness assessments for AI systems.