MLOps Belgium meetup, December 2022

December 2022 edition of MLOps Belgium meetup!

Event will be at the In The Pocket offices but a link will be shared so that you can follow remotely.

Pizza and drinks will be available between 18h and 19h.

Also, don't register if you are not able to attend physically. A link will be provided to follow remotely but we want to use the attendance list to estimate the number of people being present for food and drinks. Please mind the organisational effort from the people involved. Thanks!

Overview of the talks:

Hi ML meet UX
Speaker: Rémi Delanghe, In The Pocket

As AI matures and becomes democratised, we are seeing an exciting shift towards a more user-centric application of data science and Machine Learning.

Creating products with ML-technologies at their core poses some interesting challenges towards a user’s trust and expectations. That’s why we have to continuously measure how much the user trusts the results of our solution. Through experiments we verify whether the model’s performance matches the user’s expectations.

This interaction between Machine Learning & User Experience requires an iterative process with the flexibility to make big changes in the technical setup while constantly validating with users.

In this talk we will discuss how we tackle these challenges in our discovery process to improve our ML products.

MLOps for independent research, or how to run experiments on a budget
Speaker: Alexandru Burlacu

Not everyone is lucky enough to get into leading AI research labs with virtually infinite compute. And even if you do, it's not guaranteed you'll be working on what your heart desires. So how do you do deep learning research when you're budget constrained, so much so that you can't even buy a decent GPU and have to rely on the cloud? How do you increase resource utilization, reduce idle time, and ensure your experiments are replicable? Find out the answer to these questions and a few more in my presentation.

Bringing similarity-based ML models from a prototype to production
Speaker: Hannes Hapke, Digits

In this session, Hannes Hapke is sharing valuable lessons learned from bringing similarity-based ML models from a prototype to production, processing 60 million transactions per day. He will introduce the key concepts of similarity-based machine learning, address the challenges of integrating such models into products and share the lessons learned by the ML team at Digits.