

Productionising real-world ML data pipelines (Interview)
Feb 14, 2020
Yetunde Dada, a product manager at QuantumBlack and Sundance New Frontier Lab fellow, dives into the revolutionary Kedro project. She explains how Kedro transforms Python data pipelines, making them scalable and reproducible. The conversation also covers the intricacies of transitioning to open-source, user engagement, and the challenges faced along the way. In a delightful twist, Yetunde shares insights about her VR film 'Atomu,' which explores gender norms in Kenya, showcasing a blend of technology and creativity.
AI Snips
Chapters
Transcript
Episode notes
Kedro's Impact
- Kedro structures data science code, improving reproducibility and deployability.
- Open-sourcing Kedro enabled client use, community growth, and reusable analytics code.
Kedro's Foundation
- Kedro, built upon Cookie Cutter Data Science, promotes standardized workflows.
- This consistency enables self-documenting code and simplifies collaboration.
Getting Started with Kedro
- Start a Kedro project by running
kedro new
to generate a project template. - Execute
kedro run
to run your first Kedro pipeline.