AI与代理AI风险管理高级专员
AI and Agentic AI Risk Management Senior Specialist
关于Nu
Nu是拉丁美洲领先的数字银行，为巴西、墨西哥和哥伦比亚的1.4亿客户提供服务。公司通过利用数据和专有技术，开发创新产品和服务，引领行业变革。
以对抗复杂性并赋能人们为使命，Nu为客户提供完整的金融旅程，通过负责任的贷款和透明度促进金融准入和进步。公司由高效且可扩展的商业模式驱动，结合低成本服务与不断增长的回报。
Nu的影响已获得多项奖项的认可，包括《时代》100家最具影响力公司、《快公司》最具创新力公司以及《福布斯》全球最佳银行。
访问我们的机构页面 https://www.nu.com/2026-en
关于团队
在Nubank，我们高度依赖数据、机器学习，并越来越多地依赖生成式和代理式AI来推动我们的战略，为客户提供最佳体验和产品。模型风险团队在确保与我们的模型和AI系统相关的风险被理解和控制方面发挥着关键作用。我们现在正在构建一个专门的AI风险管理能力，以应对先进AI带来的新兴风险——包括基于大语言模型和自主代理系统的风险，重点关注AI质量、模型和代理行为，以及保障这些系统在内部和面向客户使用场景中的安全性和可靠性。
关于职位
这是一个高级且注重实践的技术岗位。你将帮助定义Nubank在AI和代理式AI中的模型风险管理——不仅继承现有的框架，还要构建和增强它们。你将对AI系统进行独立评估，评估其质量、行为和稳健性，并协助设计规范其安全使用的护栏和平台级控制措施。你将作为一线工程和AI开发团队的可信技术同行，提供关于AI风险的实用指导，同时不阻碍负责任的创新。该职位聚焦于AI质量、代理行为和平台控制；网络安全、数据隐私和欺诈相关事项由其他职能团队负责，不在本职位范围内。
你将负责的工作
AI风险框架与治理
- 构建并持续优化AI和代理式AI系统的风险管理框架，包括库存标准、评估方法、控制设计和问题管理。
查看英文原文
ABOUT NU
Nu is the leading digital bank in Latin America, serving 140 million customers across Brazil, Mexico, and Colombia. The company has been leading an industry transformation by leveraging data and proprietary technology to develop innovative products and services.
Guided by its mission to fight complexity and empower people, Nu caters to customers’ complete financial journey, promoting financial access and advancement with responsible lending and transparency. The company is powered by an efficient and scalable business model that combines low cost to serve with growing returns.
Nu’s impact has been recognized in multiple awards, including Time 100 Most Influential Companies, Fast Company’s Most Innovative Companies, and Forbes World’s Best Banks.
Visit our Institutional Page https://www.nu.com/2026-en
About the team
At Nubank we heavily rely on Data, Machine Learning, and increasingly on Generative and Agentic AI to drive our strategy and deliver the best experience and products to our customers. The Model Risk team plays a crucial role in ensuring the risks associated with our models and AI systems are understood and under control. We are now building a dedicated AI Risk Management capability to address the emerging risks of advanced AI — including LLM-powered and autonomous agentic systems — with a focus on AI quality, model and agent behavior, and the platform controls that keep these systems safe and reliable across internal and customer-facing use cases.
About the role
This is a senior, hands-on technical position. You will help define what model risk management looks like for AI and Agentic AI at Nubank — building and enhancing the frameworks, not just inheriting them. You will perform independent assessments of AI systems for quality, behavior, and robustness, and help design the guardrails and platform-level controls that govern their safe use. You'll act as a credible technical peer to first-line engineering and AI development teams, providing practical guidance on AI risk without slowing responsible innovation. This role focuses on AI quality, agentic behavior, and platform controls; cybersecurity, data privacy, and fraud-specific matters sit with partner functions and are out of scope.
What you will do
AI Risk Framework & Governance
- Build and continuously enhance the risk management framework for AI and Agentic AI systems, including inventory standards, assessment methodologies, control design, and issue management.
- Inventory and map Nubank's AI use cases to surface gaps, materiality, and the most critical risks, and define prioritized mitigation actions.
- Assess whether first-line monitoring is effective, proportionate to model risk, and sufficient to keep AI systems fit for purpose over time.
Independent AI Assessment
- Perform independent technical assessments across generative AI, and agentic workflows, covering data, assumptions, methodology, testing, behavior, and monitoring.
- Assess risks in LLM-powered applications, including RAG pipelines, tool use, autonomy boundaries, model/agent quality, human oversight, and hallucination risk.
- Identify and document model limitations, failure modes, and emerging AI risks, including drift, instability, fairness, and robustness concerns.
Controls, Platform & Enablement
- Influence first-line teams on platform architecture and embedded controls for the safe deployment and monitoring of AI.
- Build Key Risk Indicators (KRIs) and metrics for continuous monitoring of AI risk.
- Develop tools, evals, analyses, and playbooks (including AI-enabled automation) to improve the speed, scale, and effectiveness of AI governance and validation.
Advisory & Advocacy
- Serve as a trusted advisor across the AI/ML lifecycle, evaluating new use cases for materiality and governance requirements prior to deployment.
- Discuss and report AI risk status and independent opinions to stakeholders, including senior managers and, where relevant, regulators.
- Champion AI risk management as a strategic enabler of safe and scalable AI adoption, and build AI risk literacy across engineering, product, and risk teams.
- Work in a multicultural, diverse, and highly skilled environment.
Requirements we are looking for
- Education: A bachelor's or master's degree in a quantitative field (computer science, data science, statistics, mathematics, engineering, or related).
- Hands-on AI/ML experience: A track record developing or validating AI/ML models and systems, ideally a candidate who has moved from an AI / Machine Learning Engineer background into model risk, governance, or risk management. You don't need to have trained foundation models from scratch, but you need solid, current technical depth.
- Strong technical foundations: Proficiency in Python, SQL, and modern ML tooling; familiarity with LLMs, RAG systems, prompt engineering, and AI agent frameworks.
- Evaluation and testing: Experience evaluating and testing ML and generative AI systems, including custom evals, benchmarking, stress testing, and drift/degradation monitoring.
- Risk management experience: Demonstrated experience in risk identification, control definition, and framework building; understanding of model risk governance principles and independent effective challenge.
- Data skills: Experience working with large datasets and building dashboards and analyses to support risk visibility.
- High agency and adaptability: Comfortable operating in ambiguity, synthesizing fragmented technical and business context into a clear view of how complex AI systems actually work, and making sound judgments without a complete playbook.
- Influencing skills: Able to engage and align stakeholders across first and second lines of defense as a credible technical peer.
- Communication: Strong written and verbal skills, you can explain AI risk to a data scientist and to a regulator, and use different language for each.
- Advanced or fluent English: You will meet with partners and stakeholders across countries and prepare documentation and presentations in English.
- PLUS: Experience in a 2nd or 3rd line of defense.
- PLUS: Familiarity with regulatory Model Risk Management and AI frameworks (e.g., SR 11-7 / SR 26-2 / OCC 2011-12, NIST AI RMF, EU AI Act).
Total compensation includes base salary, RSUs and benefits. Base salary range: US$108k - US$131k.
Benefits
- Opportunity of earning equity at Nu
- Medical Insurance
- Dental and Vision Insurance
- Life Insurance and AD&D
- Extended maternity and paternity leaves
- Nucleo - Our learning platform of courses
- NuLanguage - Our language learning program
- NuCare - Our mental health and wellness assistance program
- Extended maternity and paternity leaves
- 401K
- Saving Plans - Health Saving Account and Flexible Spending Account
- Work-from-home Allowance
- Relocation Assistance Package, if applicable.
WORK MODEL FOR THIS ROLE
Hybrid 2-3 times/week: Our hybrid work model brings us to the office at least twice a week, on strategic days designed to maximize team connection and collaboration.
For more details, visit https://building.nubank.com/nu-hybrid-work-model/
Our recruitment process may involve the use of artificial intelligence–enabled tools, such as automated interview transcription and analysis, to support the evaluation process. Artificial intelligence is not used to make final hiring decisions; all decisions are made by human reviewers.