Virologist & AI Safety Evaluator (Remote Contract)

10a Labs; United States; A distancia; USD 120.000 - 180.000; Jornada completa

Descripción de la vacante

10a Labs is seeking a Virologist to design and execute rigorous evaluations of advanced AI systems, including LLMs, for biological knowledge, safety, and security implications. The role focuses on developing benchmarks, conducting red-teaming, and communicating findings to multidisciplinary teams.

The candidate will work with scientists, engineers, and AI researchers to assess capabilities and potential misuse risks in a remote United States setting, with in-person onboarding considerations as

Formación

• PhD or equivalent experience in virology or related biological science.
• Experience with pathogenic viruses, bacteria, or similar systems.
• Strong knowledge of biological research methods and experimental design.
• Understanding biosecurity and dual-use research risks.
• Experience critically evaluating scientific literature and methodologies.
• Excellent analytical and communication skills for multidisciplinary audiences.

Responsabilidades

• Design benchmarks and evaluation methodologies for AI models' biological knowledge.
• Conduct model evaluations and red-teaming to assess misuse risks.
• Develop scenarios and tasks across virology and microbiology domains.
• Assess AI-generated outputs for technical accuracy and biosafety implications.
• Identify misuse scenarios involving biological agents and dual-use tech.
• Contribute to evaluation criteria, rubrics, and measurement methods.
• Collaborate with AI researchers and engineers to implement scalable evaluations.
• Analyze results and contribute to technical reports and client deliverables.

Conocimientos

Biological analysis

Analytical thinking

Scientific communication

Educación

PhD in Virology, Microbiology, Molecular Biology or related field

Herramientas

Python

Command-line tools

Descripción del empleo

10a Labs is seeking a Virologist to design and execute rigorous evaluations of advanced AI systems, including LLMs, for biological knowledge, safety, and security implications. The role focuses on developing benchmarks, conducting red-teaming, and communicating findings to multidisciplinary teams.

The candidate will work with scientists, engineers, and AI researchers to assess capabilities and potential misuse risks in a remote United States setting, with in-person onboarding considerations as