THE OPPORTUNITY
A role worth
a closer look.
Help conversational agents handle enterprise work safely in real deployments. Investigate failures involving tool use, sensitive data and prompt injection, then build adversarial tests and safeguards that address them. Work across research, security and product teams to evaluate interventions and deploy more reliable agent behaviour.
Experience highlighted by the employer
- At least two years in AI or ML engineering, research or safety, including hands-on language-model or agent work.
- Python fluency and experience with adversarial testing, post-training methods or production safeguards.
Location & eligibility
Locations are San Francisco and New York City. The description identifies Decagon as an in-office company. Confirm work authorization and sponsorship directly; no sponsorship promise was found in the reviewed text.
Compensation
USD $200,000–$400,000 / year. Employer-published annual range plus equity; final compensation can depend on experience and US location.
Before you apply
This is TaskBull’s concise summary, not the employer’s full job description. Review the complete responsibilities, terms, accessibility information and privacy notice on the original application page.
Read the full employer listing ↗