Lead Research Scientist

This is a fully remote position, open to applicants in Singapore.

📋 Description

• Establish and advance Neo’s research agenda, determining where researchers, engineers, computational resources, and external partnerships can make the most significant impact.

• Direct major research initiatives from the selection of problems and methodologies to analysis and publication.

• Create rigorous strategies to investigate challenging-to-observe frontier-model behaviors, including incentives for evaluation recognition and concealment.

• Provide guidance and mentorship to researchers, evaluate experimental designs and outcomes, and uphold a high standard for evidence.

• Collaborate closely with research engineers to transform ambitious research inquiries into solid experimental systems.

• Foster partnerships with AI Safety Institutes, leading frontier laboratories, academia, and model developers, while representing Neo’s research externally.

• Assist in recruiting outstanding researchers and shaping Neo’s research culture as the organization expands.

• Identify unique questions where Neo can contribute distinctively and develop research to answer them convincingly.

⛳️ Requirements

• Proven track record of original research in AI safety, model evaluations, machine learning, or a related discipline.

• Strong comprehension of frontier-model behavior and evaluation or elicitation methodologies.

• Ability to translate broad issues into tangible research programs and publish results effectively.

• Exceptional technical writing skills and communication proficiency with researchers, engineers, and external partners.

• Comfortable operating in an early-stage environment where key questions may not yet be fully resolved.

• Collaborative working style with a readiness to remain hands-on.

• PhD or equivalent research output is advantageous, but not mandatory.

• Research experience in dangerous capabilities, alignment, agentic systems, or model evaluations is beneficial but not required.

• Experience in leading research projects or developing a research function is helpful but not necessary.

• Familiarity with frontier safety reports, the EU AI Act, or NIST AI RMF is useful but not essential.

• Proficiency in reading or writing Mandarin is advantageous but not required.

• Remote team members should be able to maintain some overlapping working hours with colleagues across different regions.

🏝️ Benefits

• Competitive benefits and leave policies.

• Dedicated research budget and computing resources.

• Expanding network of external collaborators.