AI Research Scientist
Join our research team to investigate vulnerabilities in frontier AI models and develop new methods for detecting and preventing adversarial attacks on autonomous systems.
About the position
As an AI Research Scientist at SnowCrash Labs, you will work at the cutting edge of AI security research. Your work will directly shape how organizations understand and defend against the security risks posed by autonomous AI agents. You will design and run experiments to uncover new categories of vulnerabilities, develop mathematical frameworks for measuring risk, and publish findings that advance the field.
Job responsibilities
- Research and catalogue new vulnerability categories in frontier AI models, expanding our proprietary weakness enumeration framework.
- Design experiments to test how AI agents respond to adversarial inputs across single-turn and multi-turn interaction scenarios.
- Develop techniques for monitoring the internal cognitive state of AI models to detect deception, bias, and manipulation before they appear in outputs.
- Collaborate with the engineering team to translate research findings into production-ready detection and prevention capabilities.
- Contribute to published research and thought leadership that positions SnowCrash Labs as the authority in AI security.
This role is ideal for someone who thrives at the intersection of security research and machine learning, and who wants their work to have immediate real-world impact on how enterprises deploy AI safely.
Benefits
Lorem ipsum dolor sit amet consectetur nec quis suspendisse nulla amet viverra tortor.
