Product Manager, Safety Research
Core
Bridge AI safety research and the North product to translate model behavior findings into concrete safety features, guardrails, and controls.
Role type
Product Manager, Safety Research
Builds
Safety guardrails, controls, and safeguards for the North agentic AI platform
Domain
Enterprise AI, Agentic Systems, AI Safety
Deliverable
product features
Required skills
Product management, Research operations, AI safety, Model behavior analysis, Evaluation framework design, Cross-functional collaboration, Strategic prioritization
Preferred skills
LLM evaluation, Red-teaming, Prompt injection defense, RAG poisoning mitigation, Zero-to-one process building, Agentic AI systems experience
Technologies
LLMs, Agentic AI platforms, Safety evaluation frameworks
Responsibilities
Translate model evaluation and red-teaming findings into product-level guardrails; Own the safety product roadmap based on research and threat vectors; Partner with modeling teams to interpret safety evaluations across adversarial inputs; Define evaluation frameworks to assess safety properties as models evolve; Coordinate development of intervention mechanisms across research, engineering, and policy; Monitor emerging misuse patterns in agentic systems; Build processes for scaling safety review before feature launches
Seniority
Mid-Senior, hands-on IC
