Learn MLflow for experiment tracking, model registry, model deployment, prompt management, and Databricks AI Functions
Machine learning projects often start as simple notebooks, but as teams grow and models move toward production, managing experiments, models, and deployments becomes difficult.
What you’ll learn
- Understand how MLflow works internally and how it fits into real MLOps workflows for experiment tracking, model lifecycle management, and deployment..
- Track machine learning experiments using MLflow by logging parameters, metrics, artifacts, and runs in a structured and reproducible way..
- Build and manage ML models using MLflow Model Registry including versioning, lineage tracking, and production model management..
- Deploy ML models as REST APIs using MLflow’s built-in model serving capabilities for real-time inference..
- Implement LLMOps workflows using MLflow including prompt registry, prompt versioning, evaluation, and prompt management..
- Integrate MLflow with Databricks to manage machine learning experiments and production ML pipelines..
- Use Databricks AI Functions to perform tasks like sentiment analysis, classification, text extraction, and schema extraction using SQL..
- Build an end-to-end ML workflow including experiment tracking, model logging, model registry, and deployment..
Course Content
- Introduction –> 1 lecture • 2min.
- Mlops using mlflow –> 9 lectures • 1hr 20min.
- LLMOps using mlflow –> 8 lectures • 2hr 16min.
- Databricks – Mlflow –> 2 lectures • 1hr 49min.
- Databricks AI functions –> 9 lectures • 1hr 1min.
Requirements
Machine learning projects often start as simple notebooks, but as teams grow and models move toward production, managing experiments, models, and deployments becomes difficult.
How do teams track experiments?
How do they manage model versions?
How do they deploy models reliably?
And how do modern teams manage LLM prompts and GenAI workflows?
This is where MLflow comes in.
In this course, you will learn how MLflow is used in real-world MLOps systems to manage the entire machine learning lifecycle.
Instead of focusing only on APIs, this course explains the system-level thinking behind MLflow so you can understand how ML systems are built in production environments.
What You Will Learn
By the end of this course, you will understand how to:
• Track machine learning experiments using MLflow
• Log parameters, metrics, artifacts, and runs
• Use MLflow Model Registry to manage model versions
• Deploy models using MLflow model serving
• Understand backend store and artifact store architecture
• Implement nested runs for advanced experiment tracking
• Use MLflow for LLMOps workflows including prompt registry
• Evaluate prompts and manage prompt versions
• Integrate MLflow with Databricks workflows
• Use Databricks AI Functions for AI-powered SQL tasks
Practical Learning Approach
This course focuses on hands-on demonstrations.
You will learn how to:
- Set up MLflow from scratch
- Track experiments locally
- Understand MLflow’s internal architecture
- Log and manage machine learning models
- Deploy models as REST APIs
- Build prompt management workflows for LLM applications
- Use MLflow together with Databricks
Who This Course Is For
This course is ideal for:
- Machine Learning Engineers
- Data Scientists
- MLOps Engineers
- Developers transitioning into AI infrastructure roles
- Anyone who wants to understand how ML systems operate in production
Requirements
Basic knowledge of:
- Python programming
- Machine learning concepts
- Command line basics
No prior MLflow experience is required.
By the end of this course, you will have a clear mental model of MLflow and modern MLOps workflows, helping you build and manage machine learning systems more effectively.