Similar Jobs
See allMember of Technical Staff
AI Digest
Global
Data Analysis
Software Engineering
AI Safety
Associate Program Officer
Centre for Effective Altruism
US
AI Safety
AI Governance
Analytical Skills
Head of AI Research
A1
Global
Python
PyTorch
JAX
AI Safety Expert — English & Swedish
Partner Company
Canada
Cybersecurity
Penetration Testing
Creative Writing
AI Partner
Convergent Research
US
Fundraising
Board Governance
Responsibilities:
- Own research projects end-to-end, from hypothesis and experimental design through analysis and publication.
- Develop rigorous evaluations for misalignment, loss of control, and harmful manipulation risks.
- Study behaviors such as evaluation awareness, sandbagging, and dishonesty in agentic settings.
Requirements:
- Track record of original research in AI safety, evaluations, or a closely adjacent field.
- Deep familiarity with frontier model behavior and elicitation methodology.
- Strong technical writing and ability to collaborate closely with engineers.
Logistics:
- Role can be based in Singapore or remotely, with global team overlap.
- Compensation range: $150,000 – $200,000+, with possible adjustments for exceptional candidates.
- Competitive benefits and leave policies.
Neo Research
Neo Research is an independent AI safety research organization based in Singapore that studies frontier risks in increasingly capable AI systems, with a focus on open-weight models. The team is small, offering substantial freedom to pursue important research questions and a collaborative environment with strong engineering support and external relationships.