Wednesday, 29 April 2020 | 1.00-1:45pm (AEST) inc 15 min Q&A
An Introduction to Machine Learning Feature Stores
Organisations have been investing more and more into Machine Learning capabilities but are struggling to scale their Machine Learning workflows. Although ad-hoc feature engineering and training pipelines are a quick way for Data Scientists to experiment with machine learning models, such pipelines have a tendency to become complex over time and end up assembling a huge tech debt.
In this talk we will discuss:
- The challenges of scaling Machine Learning Model
- What a feature Store is and how it helps in scaling ML models
- What are some basic considerations while designing or choosing a feature store
- We will also see why business should start thinking about implementing feature stores to stay ahead
Speaker: Harmeet Kaur Sokhi, Senior Consultant - Data Engineer, ThoughtWorks
Harmeet is a data engineer at ThoughtWorks Australia with vast experience working in cloud and data. She is passionate about data engineering and has helped in developing various data engineering solutions. She has special interest in streamlining core engineering practices in the data world and also in designing Machine Learning architecture that makes Productionising ML easy to implement and scale.