We talked about:

• Marysia’s background
• What data-centric AI is
• Data-centric Kaggle competitions
• The mindset shift to data-centric AI
• Data-centric does not mean you should not iterate on models
• How to implement the data-centric approach
• Focusing on the data vs focusing on the model
• Resources to help implement the data-centric approach
• Data-centric AI vs standard data cleaning
• Making sure your data is representative
• Knowing when your data is good enough
• The importance of user feedback
• “Shadow Mode” deployment
• What to do if you have a lot of bad data or incomplete data
• Marysia’s role at PyData
• How Marysia joined PyData
• The difference between PyData and PyCon
• Finding Marysia online

Links:

• Embetter & Bulk Demo: https://www.youtube.com/watch?v=L---nvDw9KU

Free data engineering course: https://github.com/DataTalksClub/data-engineering-zoomcamp

Join DataTalks.Club: https://datatalks.club/slack.html

Our events: https://datatalks.club/events.html