As a Senior AI Guardrail Evaluation Expert & Project Manager, you will play a pivotal role in safeguarding the integrity and trustworthiness of our AI systems. You will lead the charge in evaluating and refining the guardrails that protect our models from generating harmful, biased, or unfaithful content. This is a unique chance to combine your deep technical expertise with your leadership skills, working within a dynamic, hybrid environment to define best practices in AI ethics and safety for the financial industry.

You’ll do

• Lead the comprehensive evaluation and continuous refinement of AI guardrails to ensure robust detection and mitigation of harmful content and behaviours.

• Develop and improve sophisticated strategies for evaluating guardrails against:

• Profane, Toxic, and Adult Content

  • Bias detection and mitigation in AI outputs

  • Illegal and Harmful Content generation

  • Misuse of Personal Data (PII) and other sensitive information

  • Hallucination and the generation of unfaithful or factually incorrect content

  • Drive the planning, execution, and meticulous documentation of complex AI guardrail evaluation projects, ensuring adherence to timelines, resources, and quality deliverables.

  • Identify, onboard, train, and motivate a team of skilled human annotators, providing clear guidance and fostering a high-performance environment.

  • Leverage strong linguistic understanding to guide annotation efforts, refine guardrail performance, and ensure high-quality human feedback.

  • Collaborate cross-functionally with AI researchers, engineers, product managers, and legal teams to integrate ethical considerations throughout the AI development lifecycle.

You'll bring

• A proven track record in AI/ML ethics, content moderation, linguistic analysis, or a related field, with significant project leadership experience.

• Deep practical expertise in evaluating and improving AI guardrails on an evaluation platform like Braintrust, specifically addressing the categories mentioned above.

• Proficiency with version control systems (e.g., GitHub), collaboration platforms (e.g., Confluence), and data science environments (e.g., Jupyter Hub).

• Scripting and automation skills, particularly with Python, for data analysis and workflow optimization.

• Demonstrated ability to plan, execute, and meticulously document complex projects, managing timelines, resources, and deliverables effectively.

• Exceptional leadership skills, with experience in building, motivating, and managing high-performing teams, especially human annotators.

• Superior linguistic understanding and analytical capabilities to guide complex annotation tasks and interpret nuanced AI behaviours.

• Excellent communication, interpersonal, and problem-solving skills, with the ability to articulate complex technical concepts.

ABOUT GOLDMAN SACHS:

At Goldman Sachs, we commit our people, capital and ideas to help our clients, shareholders and the communities we serve to grow. Founded in 1869, we are a leading global investment banking, securities and investment management firm. Headquartered in New York, we maintain offices around the world.

We believe who you are makes you better at what you do. We're committed to fostering and advancing diversity and inclusion in our own workplace and beyond by ensuring every individual within our firm has a number of opportunities to grow professionally and personally, from our training and development opportunities and firmwide networks to benefits, wellness and personal finance offerings and mindfulness programs. Learn more about our culture, benefits, and people at GS.com/careers.

We’re committed to finding reasonable accommodations for candidates with special needs or disabilities during our recruiting process. Learn more: https://www.goldmansachs.com/careers/footer/disability-statement.html

© The Goldman Sachs Group, Inc., 2023. All rights reserved.

Goldman Sachs is an equal opportunity employer and does not discriminate on the basis of race, color, religion, sex, national origin, age, veterans status, disability, or any other characteristic protected by applicable law.