Staff ML Engineer - AI Safeguards & Safety Eval

Servicenow; Mountain View (CA); On-site; USD 210,000 - 330,000; Full time

Job summary

ServiceNow is seeking a senior ML engineer to design and ship safeguards for AI agents, covering prompt injections, jailbreaks, and sensitive content handling. You will build and evaluate guardrail models and classifiers, working with our Security team to implement robust mitigations in production environments.

The role requires extensive Python experience, deep knowledge of LLM safety, and the ability to balance performance with safety across enterprise and government deployments.

Qualifications

• 8+ years building production software, including 3+ years of applied ML
• Hands-on experience developing safeguards for LLMs
• Strong model evaluation fundamentals for agent trajectories and text classification
• Knowledge of how LLM safeguards get bypassed, including indirect prompt injection
• Excellent Python experience

Responsibilities

• Design, build and ship safeguards in the agent harness for prompt injection, single and multi-turn jailbreaks, toxicity and sensitive topics, and sensitive data exposure
• Build evals that measure safeguard efficacy on both false and true positives, using datasets from customer reports, adversarial testing and production traffic
• Work closely with our Security team: turn their adversarial findings into eval cases and fixes, and build mitigations together that hold when detection misses
• Select, integrate and fine-tune detection models within a stack shared by our commercial and government environments
• Balance quality, latency and cost for controls that run in the request path
• Build configurable safeguard policies so customers can decide how sensitive topics are handled
• Monitor safeguard behavior in production and fix regressions

Skills

Python

ML deployment

LLM safeguards

Content classifiers

Evaluation datasets

Job description

ServiceNow is seeking a senior ML engineer to design and ship safeguards for AI agents, covering prompt injections, jailbreaks, and sensitive content handling. You will build and evaluate guardrail models and classifiers, working with our Security team to implement robust mitigations in production environments.

The role requires extensive Python experience, deep knowledge of LLM safety, and the ability to balance performance with safety across enterprise and government deployments.