Podcast
Building Data Products at Scale: Recommenders, Domain Ownership, and Hiring for Production ML
Open original DataTalks.Club episode
Building Data Products at Scale: Recommenders, Domain Ownership, and Hiring for Production ML
Original Episode
Use these links for the canonical episode and media sources.
- Open the original DataTalks.Club podcast page
- Watch on YouTube
- Listen on Spotify
- Listen on Apple Podcasts
Episode Overview
How do you scale recommender systems, hire for production ML, and model price markdowns to reduce waste—and who should own those decisions? In this episode, Anna Hannemann, Domain Owner for Data Science at Metro.digital, walks through practical answers informed by her PhD in Data Science and prior leadership of recommender and robotics/smart logistics teams.
People
Use these links to connect the episode to guest notes.
Chapter Summary
Use these checkpoints to decide whether to open the source transcript.
- 0:00 - Episode Introduction
- 1:32 - Guest & METRO overview and customer data completeness
- 4:49 - Anna’s academic and career background (PhD, web science, logistics)
- 12:49 - Value of technical expertise for data product leads
- 15:11 - Core product owner responsibilities and team advocacy
- 20:00 - Role comparison: product owner versus product manager
- 22:08 - Recommender systems at METRO: API-first design and scaling
- 30:01 - Hiring strategy for production ML: data scientist, ML engineer, MLOps
- 34:53 - Recommender algorithms: collaborative filtering and Word2Vec variants
- 35:55 - Essential skills: metrics, trade-offs, and technical literacy for product
- 38:32 - Domain owner role: aligning data scientists across product teams
- 40:01 - People management at scale: directs, reviews, and cross-team enablement
- 41:34 - Price markdown modeling: reducing waste and optimal discounting
- 44:48 - Sourcing problems from operations: business-driven prioritization
- 45:57 - Managing multiple data domains: delegation, rotations, and budget ownership
- 48:44 - Evaluating new domains: MVPs, manual fixes, and business justification
- 53:09 - Portfolio approach: validating and staging data product investments
- 54:21 - Community leadership: organizing ProductTank meetups
- 57:48 - Recommended resource: “Data Science for Business” for data product roles