Collecting the three event families (and keeping the heartbeat), choosing real-time vs batch deliberately, running a feature store that serves the same numbers in training and production, getting point-in-time correctness right, and enforcing the viewing-data privacy boundary (VPPA, GDPR, CCPA) — on a single sheet.
Download free PDF
The handbook · 7 volumes
Fora Soft Video Engineering Handbook
This one-page checklist reviews event collection, the feature store, point-in-time correctness and the viewing-data privacy boundary before you build personalization. We wrote seven volumes on the video stack; start with Volume 3 for player events, QoE monitoring and delivery economics, then Volume 7 for model serving, cost per stream-hour and compliance such as the EU AI Act. Written by Nikolay Sapunov, CEO at Fora Soft.
Building a personalization data pipeline for your streaming service and unsure where to start?
Share your event schema and current data stack, and we will map the gaps and plan a pipeline that trains and serves on the same numbers. 250+ video products shipped since 2005.
You get a realistic estimate, technical recommendations and next steps. Free, no obligations.
