MLOps and Data Pipeline Orchestration for AI Systems
1h 19mIntermediate2025-05-27
Authors
Janani Ravi
Certified Google Cloud Architect and Data Engineer
Course details
This course covers the automation and management of machine learning workflows, from data ingestion to model deployment. Join instructor Janani Ravi as she teaches you how to orchestrate and optimize data pipelines, ensuring efficient, scalable, and reliable operation of AI systems in production environments. This course is an ideal fit for anyone working with AI, data infrastructure, and machine learning operations (MLOps), including data engineers, AI and ML engineers, MLOPs engineers, and DevOps engineers.
Skills covered
Machine LearningData EngineeringArtificial Intelligence FoundationsArtificial Intelligence (AI)Data ScienceOne-Off
Concepts
0. Introduction
- 01 - Importance of MLOps
- 02 - Prerequisites
1. The Need for MLOps
- 03 - Agile development and DevOps
- 04 - Introducing MLOps
- 05 - The MLOps lifecycle
- 06 - Tracking artifacts in MLOps
2. MLOps with MLflow
- 07 - Introducing MLflow
- 08 - Install MLflow and prepare data for machine learning
- 09 - Track a model run and register a model
- 10 - Multiple model versions and predictions using registered models
3. LLMOps for Large Language Models
- 11 - Introducing LLMOps
- 12 - LLMOps vs. MLOps
- 13 - LLM model development and evaluation
- 14 - LLM model deployment and operations
- 15 - Benefits, best practices, and considerations for LLMOps
4. Data Orchestration Pipelines
- 16 - Components of a data orchestration pipeline
- 17 - Detailed overview of pipeline components
- 18 - Data orchestration pipeline best practices and dos and don'ts
Conclusion
- 19 - Summary and further study