Who we are
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’s mission is to build the economic infrastructure for the internet. To accomplish this ambitious agenda, Stripe needs to enable business, banks, payment infrastructure and customers at scale. Stripe is required by law or contractual agreements (with partners) to prevent misuse of its products in the form of fraud, money laundering, terrorist financing and many financial crimes and illegitimate actions. Building trust with regulators, banks, businesses & customers is a key ingredient for Stripe to be successful and we are building best-in-class ML/AI to help deal with risk mitigation at scale, enable good businesses and develop high partner trust.
The Financial Crimes engineering team is responsible for building the platform, tooling, data architecture and ML models to help Stripe detect and action on compliance risks like Anti Money Laundering, sanctioning, KYC to name a few Our work impacts Stripe's reputation as a reliable and trustworthy company to manage the world's financial transactions and protects Stripe from legal action and punitive fines. In short we impact Stripe's bottom line and help build a safer financial backbone for the internet.
What you’ll do
Our ambition is to build industry-leading solutions to hard problems like:
- Modeling reputational risk
- Building ML models for near-real time detection of financial crimes in an adversarial and dynamically evolving environment
- Incorporating new signals and paradigms like digital-wallets for countering crypto based financial crimes and money laundering efforts
- Building multi-language ML models that work in multiple markets and regions
- Develop merchant risk profiles that can be used for multiple downstream applications like fraud, credit and supportability risks
- Building risk engineering systems at scale, to help Stripe grow its footprint rapidly
You will have an outsized impact on the direction, design & implementation of the solutions to these problems.
- Setting the technical & process direction for the team based on business goals
- Building innovative AI based solutions to implement product ideas that directly increase Stripe’s ability to detect illegal actions and actors
- Using NLP and CV modeling to extract signals from merchant websites and transaction logs and using them to build ML models that detect Money Laundering, transaction monitoring, risk scores associated with merchants to name a few
- Building systems that evaluate businesses for risk and take appropriate actions
- Build solutions for the Financial Crimes product team and other partner teams directly impact Stripe’s risk and compliance posture
- Helping engineers across the company to consume our models, score and artifacts
- Debugging production issues across services and multiple levels of the stack
Who you are
We’re looking for someone who meets the minimum requirements to be considered for the role. If you meet these requirements, you are encouraged to apply. The preferred qualifications are a bonus, not a requirement.
- Have at least 5 years of software engineering or Machine Learning experience
- Enjoy and have experience building scalable ML backend infrastructure
- Hold yourself and others to a high bar when working with production systems.
- Thrive in a collaborative environment
- Have experience with building and deploying ML models, especially leveraging NLP or CV on Tensorflow or pytorch.
- Have experience in Python, Scala (Spark), Java or Ruby
- Have experience in application of deep learning to risk problem areas including transformer based models for multi-class/multi-label predictions, fine-tuning of pre-trained NLP models, few-shot learning for adapting to constantly changing risk environments etc.