data science architecture
summary
Architecture centers on an immutable event history as the source for models, borrowing the philosophy of event sourcing and treating user events as the canonical data. The stack targets sub-second responses for web users and cuts model rollout time to two-week model-to-production cycles instead of months by combining clever data pipelines, realtime updates, and high-performance tooling such as Spark for large-scale ML and fast reporting.
AI-written summary · about
chapters (6)
- 0:00Intro
- 1:05How did this architecture come about
- 2:10Expectations
- 2:33Data architecture vs software architecture
- 3:25Data architect vs data scientist
- 3:52What projects are you following
related talks
34mTales of Data Architecture Evolution
22mData Ecosystem to power AI
49mTransforming Analytical Data with Event-Driven Architecture
44mWhat's the Best Big Data Architecture for You?
09mData science is no longer a back office function
43mData Modelling for Startups: Domain Models & the Modern Data Stack
01h 17mData Engineering in the AI Era: Building a Robust Foundation for AI Applications
08mDesigning Scalable, Secure Data Architectures for AI at Enterprise Scale