Content Red Team LLM Portuguese

#LI-EU

We are seeking a highly motivated and detail-oriented Policy Tester/ Prompt Generation Associate to join our Prompt Generation testing team responsible for ensuring platform/product integrity and security to fight abuse across product(s) enabled by AI/ GenAI.

Liaise and collaborate with internal and external stakeholders such as policy experts, GenAI/ AI experts, product/ engineering experts, cyber security experts for testing requirement gathering and execution, and cross-functional project management. In this role, candidate will be responsible for supporting testing plan development, executing test plans, gathering test results and support developing vulnerability assessment reports detailing potential loopholes and issues found during the testing.

Responsibilities:

• Identify potential policy violations and unintended responses that violate the given policy guidelines to help improve the product(s) by generating multiple prompts for testing;
• Provide effective support to internal and external stakeholders for large-scale model projects, including text, images, video, audio, or graph generation in line with the testing scope of work;
• Collaborate with resources across T&S, GenAI /AI, and product teams to generate prompts and develop comprehensive test reports, which include, but are not limited to, vulnerability assessments and potential solutions wherever applicable;
• Maintain clear communication with internal and external stakeholders. Engage with testing teams, coordinate test results and reporting requirements internally, and ensure orderly progress of test execution tasks;
• Manage end-to-end prompt generation for model training to ensure model quality, robustness, and safety within the given policy guidelines.

Qualifications:

• >1 years of experience in content moderation, prompt engineering, policy testing, anti-abuse, cybersecurity, or related fields;
• 12th Grade (completed) or higher - mandatory;
• Interest for GenAI/ ML/ AI/ new technologies/cybersecurity and experience working with similar / related industry (Familiarity with latest trends, challenges, and opportunities in generative AI security and ethics is a plus);
• Familiarity with content management frameworks and optimization. Passion for applying these skills to human/AI alignment;
• Fluent in English & Portuguese. Excellent verbal and written communication skills with ability to collaborate with stakeholders at various levels effectively;
• Self-starter with a passion for continuous learning and experimentation;
• Comfortable working in a fast-paced, results-driven environment with an eye to detail and accuracy;
• Ability to operate in areas that are not clearly defined within policy;
• Ability to prioritize and multitask to meet various client test requirements.