Generative AI Safety & Model Validation Lead

TikTok USDS Joint Venture; San Jose (CA); On-site; USD 187,000 - 438,000; Full time

Job summary

TikTok USDS Joint Venture is seeking a Model Validation Lead to build a scalable safety and security validation program for all generative AI models. You will set the technical direction, define testing methods, and lead complex engagements without managing people directly.

You will partner with TnS, Penetration Testing, and Red Team to design evaluations, validate safeguards, and identify cyber threats. You will establish pipelines, acceptance criteria, and governance for model deployment and

Qualifications

• Bachelor degree in CS, Info Security, AI, or related field; 8+ years in security testing or related discipline.
• Hands-on experience evaluating large language models or multimodal generative AI for adversarial behavior or unsafe outputs.
• Ability to set technical direction, lead cross-functional programs, and mentor engineers without formal mgmt authority.
• Experience designing repeatable test plans, adversarial datasets, automated evaluations, and measurable criteria.
• Strong knowledge of application security, cloud security, offensive security, and threat modeling.
• Proficiency in Python and another language; building test harnesses and automating validation workflows.
• Able to explain complex model failures and attack paths to technical and non-technical stakeholders.

Responsibilities

• Define the technical strategy, roadmap, and operating model for model safety and security assurance.
• Establish testing methods, tooling, test infrastructure, and specialist capacity for JV models.
• Design and execute adversarial evaluations of safeguards including content controls and policy enforcement.
• Validate cyber threats from generative AI with Penetration Testing and Red Team partners.
• Lead pre-deployment and continuous testing of foundation and updated models and agents.
• Build automated evaluation pipelines and curated adversarial test suites.
• Translate failures into clear risk, remediation, and governance actions.
• Define SOPs, ROE, evidence requirements, and testing methodologies for validation.
• Produce an authoritative view of coverage, risk, and regression trends for leadership.

Skills

Generative AI Security

Technical Leadership

Evaluation Design

Security Knowledge

Python Automation

Communication

Adversarial Testing

Education

Bachelor's degree in CS/Info Security/AI or related field

Job description

TikTok USDS Joint Venture is seeking a Model Validation Lead to build a scalable safety and security validation program for all generative AI models. You will set the technical direction, define testing methods, and lead complex engagements without managing people directly.

You will partner with TnS, Penetration Testing, and Red Team to design evaluations, validate safeguards, and identify cyber threats. You will establish pipelines, acceptance criteria, and governance for model deployment and