Staff Machine Learning Model Risk Specialist

The Team:

Upstart’s Model Risk team is responsible for ensuring that the risk of models — including all models impacting the new Upstart Bank — is well-understood, monitored, and mitigated. For years, machine learning (ML) models have been the key, differentiating technology at Upstart and an exciting area of focus for the team, but we are also expanding our scope to include all modeling methodologies and Generative AI applications across the Bank. This work is essential for Upstart’s internal risk management, ensuring that our models help us make better decisions and maintain credibility with external stakeholders such as our regulators and capital providers. The team’s focus is on articulating sound model risk management principles and implementing them in collaboration with our peers on Upstart’s Risk and Machine Learning teams. This work also includes explaining our models to stakeholders, supporting external validations, and conducting analyses to reinforce our goals.

As a Staff Model Risk Specialist at Upstart, you will independently execute core components of the model risk management program supporting Upstart Bank. You will oversee risk across a diverse and growing inventory of models and Generative AI applications, including sophisticated machine learning models used in lending and other models supporting areas such as fraud, compliance, finance, capital and liquidity, servicing, and operational risk.

This presents a unique opportunity to help build a comprehensive model risk management program for a new bank. You will evaluate model and GenAI application documentation, monitoring, governance, and risk assessments while partnering with developers, business sponsors, and risk stakeholders to identify and address emerging risks. You will apply a risk-based approach across technologies that range from traditional statistical methods to advanced machine learning and GenAI systems, adapting your review to their different purposes, complexities, and risk profiles. You will also translate complex technical concepts into clear, decision-useful information for audiences with varying levels of technical expertise.

How you’ll make an impact

• Partner with Machine Learning teams, GenAI application developers, business sponsors, and other stakeholders to maintain accurate inventories, risk assessments, documentation, monitoring reports, and supporting governance materials for models and GenAI applications affecting Upstart Bank.
• Review methodologies, assumptions, data inputs, system designs, performance measures, controls, and limitations to provide effective challenge and identify areas requiring further analysis or remediation.
• Apply a risk-based approach to evaluate a broad range of quantitative methods and technologies, from traditional statistical and financial models to complex machine learning models and GenAI applications.
• Conduct and document model risk assessments, monitoring reviews, and targeted quantitative analyses that support internal policies and regulatory expectations.
• Help develop practical governance approaches for new and rapidly evolving technologies, particularly machine learning and GenAI applications for which risks, evaluation methods, and industry practices continue to evolve.
• Respond to model- and GenAI-related questions from regulators, lending partners, and other external stakeholders in collaboration with Machine Learning, business teams, Legal, Compliance, and partner-facing teams.
• Track model risk issues, remediation plans, program goals, and emerging risks, escalating material findings and recommending practical improvements as the Bank’s model inventory and use cases continue to expand.

Minimum Qualifications

• Master’s degree in quantitative field such as finance, mathematics, economics, statistics or a related discipline
• 4+ years of experience in model risk management, model validation, model governance, machine learning, data science, quantitative risk, AI governance, or a related technical risk function.
• Internship or project experience related to model risk management, model validation, machine learning, or data science.
• Basic understanding of AI/ML methodologies such as tree-based models and neural networks.
• General familiarity with GenAI applications.
• Experience coding in R, Python, or similar languages such as Matlab

Preferred Qualifications

• PhD in a quantitative field of study such as statistics, econometrics, finance, mathematics; or a related discipline
• 5+ years of experience in model risk management or model governance, or related fields such as ML and Data Science, Risk, Trust and Safety, or Technical Writing
• Familiarity with GenAI applications, including evaluation approaches, prompt and system design, retrieval-augmented generation, tool use, guardrails, and ongoing monitoring.
• Experience assessing models used outside of credit underwriting, such as models supporting fraud, compliance, finance, capital and liquidity, servicing, operational risk, or financial reporting.
• Strong communication skills: ability to adapt technical information to varying needs and audiences, and managing trade-offs such as providing modeling detail while protecting intellectual property
• Proactive mindset with the ability to take initiative
• Understanding of advanced AI/ML topics such as model monitoring, fairness, and explainability
• Advanced coding skills in R, Python, and SQL, and experience using Git
• Interest in or knowledge of consumer lending, credit risk, model fairness, explainability, or the use of machine learning and GenAI in a regulated environment.

Position location This role is available in the following locations: Remote

Travel requirements As a digital first company, the majority of your work can be accomplished remotely. The majority of our employees can live and work anywhere in the U.S but are encouraged to to still spend high quality time in-person collaborating via regular onsites. The in-person sessions’ cadence varies depending on the team and role; most teams meet once or twice per quarter for 2-4 consecutive days at a time.

#LI-REMOTE

#LI-MidSenior