DataTalks.Club From Theme Parks to Tesla: Building Data Products That Work
Oct 10, 2025
Abouzar Abbaspour, a data engineer with a rich background in machine learning and recommendation systems, shares his unique career journey from software engineering in Iran to working at Tesla. He delves into crowd modeling at Efteling, revealing how data from rides influenced visitor flow solutions. His transition to Bol.com highlights innovative brand recommendations, including a fun 'Tinder for brands' prototype. He also discusses the impact of large language models on productivity and how practical inference is revolutionizing his work at Tesla.
AI Snips
Chapters
Transcript
Episode notes
From Software To Theme-Park Crowds
- Abouzar joined Efteling to work on crowd control and was initially unsure how his skills fit an amusement park.
- He built models from surveys and transactions to predict queues and recommend next moves to visitors.
Modeling Visitor Routes From Surveys
- Efteling collected one-year route data from ~3,000 survey participants to model visitor patterns.
- They matched user preferences to common route variations and recommended the highest-probability attractions.
Building A Favorite-Brand Carousel
- At Bol.com Abouzar built a 'favorite brand' carousel to surface brands on app and web shop.
- The feature combined user behavior and brand popularity to surface relevant brands and products.