Machine Learning Engineer, Capital Underwriting

Who we are About Stripe Stripe is a financial infrastructure platform for businesses. Millions of companies—from the world’s largest enterprises to the most ambitious startups—use Stripe to accept payments, grow their revenue, and accelerate new business opportunities. Our mission is to increase the GDP of the internet, and we have a staggering amount of work ahead. That means you have an unprecedented opportunity to put the global economy within everyone’s reach while doing the most important work of your career. About the team Stripe Capital provides access to fast, flexible financing to small-and-medium businesses on Stripe to accelerate their growth, and we lent over $1B in 2024. Businesses use the funds for marketing, team growth, geographic expansion, working capital, new equipment purchases, and much more. Machine learning is core to Stripe Capital’s business—we use information about businesses from their activity within and outside of Stripe and our models to automatically underwrite uniquely tailored financing offers to their needs, which banks are often unable to do. We are doing so through models with an established performance history, data infrastructure that is Stripe scale, and a strong feedback loop that includes explainability, anomaly detection and a risk portfolio management layer. We're an end-to-end team going from ideas to models to shipping in production. What you’ll do As a machine learning engineer for Stripe Capital, you'll be responsible for designing, building, training, evaluating, deploying, and owning ML models in production with the goals of providing financing opportunities to as many users as possible while satisfying financial performance goals. You'll work closely with software engineers, data scientists, product managers, and risk managers to operate Stripe’s ML powered systems, features, and products. You'll also contribute to and influence ML architecture at Stripe and be a part of a larger ML community. Responsibilities - Design state-of-the-art ML models and large scale ML systems for underwriting and portfolio management for Stripe Capital based on ML principles, domain knowledge, risk, regulatory and engineering constraints - Design systems to speed up the time from idea to deployment of new models - Experiment and iterate on ML models (using tools such as PyTorch and TensorFlow) to achieve key business goals and drive efficiency - Develop pipelines and automated processes to train and evaluate models in offline and online environments - Integrate ML models into production systems and ensure their scalability and reliability - Collaborate with product and strategy partners to propose, prioritize, and implement new product features - Engage with the latest developments in ML/AI and take calculated risks in transforming innovative ML ideas into productionized solutions Who you are We are looking for ML Engineers who are passionate about building ML systems that touch the lives of millions. You have experience developing efficient feature pipelines, building advanced ML models, and deploying them to production. You are comfortable with ambiguity, love to take initiative, have a bias towards action, and thrive in a collaborative environment. We’re looking for someone who can bring new ideas to the table on building models able to push the state of the art at Stripe, especially within the regulatory and operational constraints of a financing business. Minimum requirements - 5+ years of industry experience building and shipping ML systems in production - Proficient with ML libraries and frameworks such as PyTorch, TensorFlow, XGBoost, as well as Spark - Hands-on experience in designing, training, and evaluating machine learning models - Hands-on experience in productionizing and deploying models at scale - Hands-on experience in orchestrating complicated data pipelines and efficiently leveraging large-scale datasets Preferred qualifications - - MS/PhD degree in ML/AI or related field (e.g. math, physics, statistics) - Proven track record of building and deploying ML systems that have effectively solved ambiguous business problems - Experience in adversarial domains such as Lending, Trading, Fraud - Experience with Deep Learning including the latest architectures such as transformers, test-time compute, reinforcement learning