ActiveFence

• Developing adversarial and risky prompt strategies across several areas of abuse to expose potential vulnerabilities in models. • Owning projects end-to-end - from initial planning and execution through quality assurance to final delivery - with accountability for outcomes. • Mentoring junior analysts and promoting a culture of knowledge exchange and continual learning within the team. • Managing extensive datasets across multiple languages and areas of abuse, ensuring precision and meticulous attention to detail. • Ongoing investigation into new tactics for circumventing foundational models' safety measures. • Partnering with cross-functional teams - engineering, product, policy - to tackle new challenges and craft forward-thinking strategies and resolutions.

• Oversee the production of datasets, reports, and analyses related to AI safety and red teaming activities. • Review and approve deliverables to ensure they meet quality, methodological, and ethical standards. • Deliver final outputs to clients following approval and provide actionable insights that address key risks and vulnerabilities. • Offer ongoing structured feedback on the quality of deliverables and the efficiency of team workflows, driving continuous improvement. • Design and refine red teaming methodologies for new Responsible AI projects. • Guide the development of adversarial testing strategies that target potential weaknesses in models across text, image, and multimodal systems. • Support research initiatives aimed at identifying and mitigating emerging risks in Generative AI applications. • Attend client meetings to address broader methodological or operational questions. • Represent the red teaming function in cross-departmental collaboration with other ActiveFence teams.

• Analyzing various content infringements to secure Generative AI tools • Collaborating with experts in diverse fields such as Hate Speech, Misinformation, Intellectual Property and Copyright • Writing adversarial prompts to identify weaknesses in AI models • Overseeing data management to guarantee the highest quality of outputs • Managing projects end-to-end, from initial planning through quality assurance to final delivery • Handling extensive datasets across multiple languages and areas of abuse • Ongoing investigation into new tactics for circumventing foundational models' safety measures • Promoting a culture of knowledge exchange and continual learning within the team

• Dive into the cutting-edge of technology, meticulously analyzing various content infringements. • Collaborate with experts in diverse fields such as Hate Speech, Misinformation, Intellectual Property, and Copyright. • Write adversarial prompts to identify weaknesses in various AI models, including LLMs, Text-to-Image, Text-to-Video, AI Agents. • Oversee data management to guarantee the highest quality of outputs. • Develop adversarial and risky prompt strategies across several areas of abuse to expose potential vulnerabilities in models. • Manage projects end-to-end, from initial planning and oversight through quality assurance to final delivery. • Handle extensive datasets across multiple languages and areas of abuse, ensuring precision and meticulous attention to detail. • Ongoing investigation into new tactics for circumventing foundational models' safety measures. • Work alongside diverse teams to tackle new challenges and craft forward-thinking strategies and resolutions. • Promote a culture of knowledge exchange and continual learning within the team.

• Own the commercial lifecycle of 8-figure enterprise accounts, accountable for renewals, upsells, and cross-sells. • Build and maintain deep relationships across multiple stakeholder levels, across Operations, Security, and Engineering departments. • Identify whitespace opportunities within existing accounts, leveraging deep subject-matter expertise to maximize the value we bring our clients. • Partner with technical teams to translate Alice’s complex model-hardening and red-teaming results into clear business outcomes and ROI for non-technical executives. • Act as the "voice of the customer" to influence solution design and execution.