Stripe Girl Geek Dinner 2017

Join Bay Area Girl Geek Dinners at Stripe for a night of “Adversarial Engineering” on Tuesday, December 12th, 2017! Hear from three Stripe engineers on various topics centered around security and machine learning. Join us at 6pm for snacks, drinks, and networking. Talks will start promptly at 6:45pm.

Talk Topic: “Explaining Decisions From Black-Box Models” by Alyssa Frazee
Machine learning models are often described as “magic” or “black boxes”: the important thing is what goes into them and what comes out, not necessarily how that output is calculated. Sometimes this is what we want: many consumers don’t really need to see inside the sausage factory, as they say—they just want a tasty tubular treat. But other times, a prediction from a box full of magic isn’t satisfying. In Stripe’s production machine learning system that declines high-risk credit card charges, businesses and customers often (rightly) want to know why our decline system made the decision it did. This talk will illustrate how we give reasons for our system’s risk judgments on payments.

Talk Topic: “Building A Simple DSL With Parslet” by Fay Wu
Learn how Stripe’s fraud detection tool evaluates rules to determine the outcome of transactions and build a simple domain-specific language using Parslet, a small Ruby library for constructing PEGs (Parsing Expression Grammars)!

Talk Topic: “Learning Representations Of Merchants” by Pamela Vagata
Come learn how we apply NLP techniques to discover structure in non-textual data and learn vector representations for Stripe’s merchants.