About the role
The team builds and runs evaluations to measure the cyber capabilities and robustness of AI models. The engineer designs new tests, analyzes data on security vulnerabilities, and helps develop systems to detect harmful model behavior.
Our summary of the posting; the original is on the Anthropic careers page.
What the posting requires
- Required
- Python
- Bachelor's degree
- Relevant field
- Evaluations
- Timelines
- Cybersecurity
- CTF
- Vulnerability research
- Exploit development
- Security research
- Preferred
- Security clearance
- AI/ML evaluation frameworks
- Offensive security
- AI security benchmarks
- Adversarial data
- Government partners
These tags are our reading of the posting. They can miss something; the original posting is the reference.