Protection Scientist Engineer (Integrity)

OpenAI

San Francisco, CA • 32m ago • Hybrid

Mid

Hybrid

220,000 - 425,000 USD/yr

  1. Select offer
  2. Add resume
  3. Analyze
  4. Improve
  5. Apply
  6. Auto-send · soon

Match your resume with similar open roles

Upload your resume to automatically match with active openings similar to this role.

PDF up to 10 MB

Your resume is private & secure

Or go to the original job offer

Potential salary for Protection Scientist Engineer (Integrity) role

Improve your resume now

$220,000 – $425,000/year

Quick Facts

  • Role: Protection Scientist Engineer (Integrity)
  • Focus areas: sextortion, trafficking, national security misuse (nation-state misuse), jailbreaks
  • Core responsibility: production harm detection and system defense

Description

Protection Scientists in Integrity design, orchestrate, and mature defense systems against usage harms enabled by OpenAI products. The role focuses on identifying abusive behavior through signals, classifiers, and pipelines, investigating adversarial/coordinated activity, and translating evolving threats into detection and interventions. You also drive cross-functional strategy, connect detection to policy and operational action, and evaluate end-to-end effectiveness as tactics change.

Responsibilities

  • Define and drive protection strategy for a harm area or product/model launch, turning ambiguous risks into priorities, roadmaps, and trade-offs.
  • Design, build, evaluate, and maintain production harm detection systems using investigative expertise and shared engineering/data platforms.
  • Develop production signals, classifiers, and data pipelines across accounts and activity; adapt them as adversaries change tactics.
  • Lead cross-functional execution by translating between specialist roles and aligning stakeholders.
  • Investigate exploitation and national security threats; recognize patterns of coercion/exploitation/coordinated misuse and convert findings into detection criteria.
  • Connect detection to effective action by partnering with Policy and Operations on review/escalation protocols and intervention improvements.
  • Evaluate and improve the full system of defense by measuring remaining harm, detection quality, and intervention effectiveness.
  • Protect launches by defining detection/review requirements, establishing monitoring, and addressing gaps through post-launch learning.
  • Scale effective approaches using language models, agents, and reusable tools with appropriate human judgment and documentation.

Requirements

  • Deep expertise in at least one relevant harm domain (sextortion, trafficking, national security, jailbreaks) or a closely related area.
  • Experience investigating adversarial behavior and translating evolving evidence into actionable detection signals.
  • Strong SQL and Python skills; experience writing reliable and efficient production code for detection and data processing.
  • Ability to use quantitative analysis and ML methods to measure risk and evaluate system effectiveness, understanding data limitations, false positives, and missed harms.
  • Experience setting strategy and delivering complex cross-functional technical projects.
  • Strong communication skills to explain findings, recommendations, uncertainty, and trade-offs.
  • Ability to use language models/automation to scale work with evaluation and human oversight.

Benefits

  • On-call rotation may include urgent escalations outside normal working hours.
  • Some investigations may involve sensitive content, including sexual, violent, or otherwise disturbing material.

See similar jobs

Explore similar jobs and use your resume score to choose where to apply next

See more

Unlock Full Resume Report