Advanced Data Processing: Batch, Real-Time, and Cloud Architectures for AI

1h 31mIntermediate2025-05-12

Authors

Kumaran Ponnambalam

Working with data for 20+ years

Course details

As AI becomes more and more integrated into enterprise applications and workflows, architecting robust and scalable AI systems becomes even more important, particularly if you’re a data scientist or AI engineer. Beyond mastering machine learning techniques and technologies, an engineer working in AI needs to be able to leverage expertise in architecting AI and ML pipelines that achieve business outcomes at scale. In this course, instructor Kumaran Ponnambalam focuses on the big picture of bringing together models, data, applications, and infrastructure to create robust architectures. By the end of this course, you’ll be prepared to design, manage, and adhere to best practices for different architecture patterns—including batch, real-time, cloud, and hybrid.

Learning objectives
Understand the unique characteristics and design constraints for different types of AI architectures.
Define the key architecture elements of AI across batch, real-time, cloud, and hybrid.
Create an architecture for a given use case by analyzing requirements and choosing the right patterns and technologies.
Scale the architecture to increase concurrency, response times, and throughput.

Skills covered

Data CentersCloud DevelopmentSpreadsheetsData EngineeringArtificial Intelligence FoundationsDatabase ManagementArtificial Intelligence (AI)Cloud ComputingData ScienceBusiness Software and ToolsOne-Off

Concepts

0. Introduction

01 - Processing data for AI

1. AI Architectures

02 - The ML life cycle
03 - Feature engineering
04 - Model training
05 - ML inference
06 - Scale and performance
07 - Architectures for AI

2. Batch AI Architectures

08 - Characteristics of batch AI
09 - Batch feature engineering
10 - Batch model training
11 - Batch Inference
12 - Scaling batch AI
13 - Batch AI architecture example - Problem
14 - Batch AI architecture example - Solution

3. Real-Time AI Architectures

15 - Characteristics of real-time AI
16 - Real-time feature engineering
17 - Real-time model training
18 - Real-time inference architectures
19 - Scaling real-time AI
20 - Real-time AI architecture example - Problem
21 - Real-time AI architecture example - Solution

4. Cloud AI Architectures

22 - Cloud and serverless computing
23 - Architecting for the cloud
24 - AI in the cloud
25 - Cloud platforms for AI
26 - Cloud AI architecture example - Problem
27 - Cloud AI architecture example - Solution

5. Hybrid AI Architectures

28 - Hybrid computing
29 - AI using hybrid computing
30 - AI architectures for hybrid computing
31 - Hybrid AI architecture example - Problem
32 - Hybrid AI architecture example - Solution

Conclusion

33 - Continuing with AI data processing