Machine Learning in Production

After a holiday break in August we're back with another PyData Amsterdam meetup! This time hosted in the new Adyen office, where of course food and drinks are served. We will hear about two cool use cases and learnings related to Machine Learning in Production. Besides, we want to try out something new: an "Alive Stack Overflow" session in which we give you the option to leverage the wisdom of the crowd to solve your current challenges. Hope to see you there.

Schedule
18:00: Welcome (🍕 & 🍺/🧃!)
19:00: [Talk 1] Nikki van Ommeren - Machine learning at Scale; Real Time Predictions in the Payment Flow
19:45: Break
20:00: [Talk 2] Sharon Gieske - Predicting Article Demand with Temporal Fusion Transformers
20:45: Alive StackOverflow - Pitch your Problem!
21:00: Networking

Talks

“Machine learning at Scale - Real Time Predictions in the Payment Flow” by Nikki van Ommeren

Adyen is the payments platform of choice for the world's leading companies, delivering frictionless payments across online, mobile, and in-store channels. Adyen processes millions of payments on a daily basis. Several machine learning models are used in an online setting to optimize the payment journey in the blink of an eye (the duration of a payment). Think of choosing the optimal payment route, deciding which data points to include in a payment request or, in case of failure, deciding on the best payment retry option. These models therefore form the backbone of the business. They have to be very fast, reliable and of course make valuable predictions. Curious how we make sure our models deliver at scale, from both an engineering and science perspective? Join us for this talk on September 7.

Nikki van Ommeren is a senior data scientist at Adyen. In the past years she has worked on various applied machine learning models, including recommendation and feedback classification models. Currently, she is working on payment optimisation using a contextual bandits approach to help merchants achieve higher payment authorisation rates. Nikki has a background in software engineering and econometrics.

“Predicting article demand with Temporal Fusion Transformers” by Sharon Gieske

Picnic is the world's fastest growing online supermarket that makes grocery shopping simple, fun, and affordable for everyone. To ensure the freshest products and reduce waste, Picnic operates as a just-in-time supply chain. This must be balanced against high availability requirements for grocery items, as one unavailable product might lead to the loss of an entire basket. Accurate article demand forecasts are paramount. In this talk, we'll share how Picnic optimizes article demand forecasts with ML models. We'll dive deeper into why we transitioned from tree-based models to deploy the more state-of-the-art Temporal Fusion Transformer model and how it's used to balance waste & availability.

Sharon Gieske is the Tech Lead of the Data Science team at Picnic. Here, she works on the development of technologies that enable Machine Learning projects in a wide range of domains such as demand forecasting, personalization, finance and more. Together with her team of data scientists, she works full-stack and builds end-to-end ML solutions to take e-commerce at Picnic to the next level.

Alive Stack Overflow
Do you have a challenge in one of your projects that you'd like input on? Pitch your problem, utilise the combined knowledge of everyone in the room to figure out your next steps! Limited to five minutes per person/challenge, but you have the opportunity to continue the discussion afterwards with a beer.

Directions
Hosted in the event space of the Adyen office at Rokin 49. Please enter via the main entrance (revolving doors) and get your visitors badge at the reception.