AI Safety Benchmarks Research Scientist

Vals; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Relocation support

Health/dental insurance

Lunch/dinner provided

Unlimited PTO

Housing stipend

Job summary

Vals AI is seeking an exceptional researcher to design and build the next generation of AI benchmarks for catching misaligned behavior in frontier models and agents. You will lead the design and development of novel benchmarks and infrastructure to assess misalignment risks in LLMs, collaborating with major labs and enterprises in the field.

You will publish findings, stay abreast of the latest alignment methodologies, and work in-person in San Francisco with relocation support to join a small,

Qualifications

• Substantial hands-on experience in safety evaluations, monitorability, scheming, deception, or multi-agent safety.
• Deep understanding of neural networks and the fundamentals of model training.

Responsibilities

• Design and develop novel, high-impact benchmarks and environments that assess real-world misbehavior in frontier models and agents.
• Collaborate with foundation model labs and enterprises to understand evolving safety evaluation needs.
• Analyze model behavior across benchmarks and articulate results.
• Publish research findings and contribute to the broader safety and evaluation research communities.
• Stay current with the latest developments in LLM capabilities and alignment evaluation methodologies.
• Location: In-person in San Francisco with relocation/transportation support.

Skills

Safety research

Deep ML/AI background

Engineering & Python

Technical communication

Tools

Python

Django

React

AWS CDK

Job description

Vals AI is seeking an exceptional researcher to design and build the next generation of AI benchmarks for catching misaligned behavior in frontier models and agents. You will lead the design and development of novel benchmarks and infrastructure to assess misalignment risks in LLMs, collaborating with major labs and enterprises in the field.

You will publish findings, stay abreast of the latest alignment methodologies, and work in-person in San Francisco with relocation support to join a small,