ML System Design - Designing Internet Scale Distributed AI/ML Systems
Details
Details
Real-world AI systems process millions of transactions per second, giving accurate predictions for billions of entities, optimized on trillion size datasets. The goal of this session is to cover the fundamental process used by top Silicon Valley companies that include multiples stages, as well as a concrete example of an ML system design.
Level - Intermediate/Advanced
Agenda
- Intro into Design & Architecture of Large Scale Distributed AI Systems
- Pre-requisites of a solid design
- 4 stages of the design
- AI/ML Architecture Blueprints
- Data science process
- Exercise/example of Designing real-world ML system
In order to get the most out of the session, it is recommended to get familiarity with the concepts defined in the pre-requisites below.
Pre-requisites
- Algorithms & Data Structures (Search, Sorting, Recursion, Trees, Graphs, DP)
- Base System Design knowledge (Microservices, Streaming, Batch, CAP Theorem, etc.)
- Base Math/ML/AI