Stripe builds the economic infrastructure for the internet. Building trust between banks, businesses & customers is a key ingredient for Stripe to be successful. Unfortunately, fraudsters & bad actors erode that trust.
The Fraud Intelligence team builds backend and machine learning (ML) systems to reduce fraud while retaining a best-in-class user experience. Our work directly impacts Stripe's bottom line, and we help build a safer financial backbone for the internet.
Here are some of the problems you will be tackling:
- How do we featurize all user interactions on Stripe and train models on high-dimensional input?
- How do we design models to handle both tabular & unstructured data?
- How do we train models to learn quickly from fresh incremental data (few-shot learning)?
- What unsupervised systems can we build to detect anomalous behavior?
You will have an outsized impact on the direction, design & implementation of the solutions to these problems.
Your work will include:
- Setting the technical & process direction for the team based on business goals
- Designing, training, improving & launching models
- Proposing and implementing ideas that directly reduce Stripe’s fraud losses
- Building systems that evaluate businesses for risk and take appropriate actions
- Leading partner teams to launch new policies that directly impact Stripe’s bottom line
- Helping engineers across the company to develop technologies for scaling our infrastructure
- Debugging production issues across services and multiple levels of the stack
You may be a good fit if you:
- Have at least 7 years of experience in training ML models
- Enjoy and have experience shipping ML models in a large-scale production environment
- Have led multiple engineers to deliver large-scale high-impact projects
- Hold yourself and others to a high bar when working with production systems
- Take pride in taking ownership and driving projects to business impact
- Thrive in a collaborative environment
- [Bonus] Have experience in Python, Scala (Spark), or Ruby