# KodeKloud: Beta Courses AWS SageMaker

## AWS SageMaker

### Machine Learning Prerequisites

- [Course Introduction](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/Course-Introduction/page.md): Introductory course teaching practical AWS SageMaker workflows including data preparation, training, model registry, deployment and monitoring with notebooks and the SageMaker Python SDK
- [Prerequisites](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/Prerequisites/page.md): Beginner prerequisites for using Python and AWS SageMaker to build, train, host, and infer machine learning models, plus basic AWS and ML workflow essentials
- [ML Basics Fundamentals of Model Training and Inference](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/ML-Basics-Fundamentals-of-Model-Training-and-Inference/page.md): Introductory guide to model training and inference using London house prices, covering data requirements, linear regression math, optimization, algorithm choices, overfitting, and deployment.
- [ML Basics Fundamentals of Model Training and Inference Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/ML-Basics-Fundamentals-of-Model-Training-and-Inference-Part-2/page.md): Intro to multivariate linear regression using a car price example, covering feature encoding, scaling, training with gradient descent, and deploying models for inference
- [ML Pipeline](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/ML-Pipeline/page.md): Overview of an end-to-end machine learning pipeline from problem framing through data prep, training, deployment, and monitoring, with Amazon SageMaker tooling for each stage.
- [How Much Math Do I Need](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/How-Much-Math-Do-I-Need/page.md): Explains practical amount of math needed for machine learning, prioritizing data preparation, basic statistics, encoding, outlier handling, scaling, and when deeper math is required
- [How Much Math Do I Need Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/How-Much-Math-Do-I-Need-Part-2/page.md): Explains min–max scaling, normalization, and standardization for preparing numerical features, when to use each technique, and scikit-learn examples.
- [How Much Math Do I Need Part 3](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/How-Much-Math-Do-I-Need-Part-3/page.md): Describes categorical encoding methods for machine learning, comparing label, one-hot, and target encoding, with advice on high cardinality, smoothing, and preventing target leakage
- [Why is SageMaker Intimidating](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/Why-is-SageMaker-Intimidating/page.md): Summarizes why SageMaker seems intimidating and how its tools align to ML roles, workflows, and deployment monitoring
- [Why Learn SageMaker by Persona](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/Why-Learn-SageMaker-by-Persona/page.md): Explains how to map Amazon SageMaker capabilities to Data Engineer, Data Scientist, and MLOps personas for reproducible, automated ML workflows, governance, deployment, monitoring, and CI CD
- [Why Learn SageMaker by Persona Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Machine-Learning-Prerequisites/Why-Learn-SageMaker-by-Persona-Part-2/page.md): Explains how data scientists perform exploratory data analysis, feature engineering, model training and iteration while leveraging managed ML platforms such as SageMaker

### SageMaker Introduction

- [SageMaker What Makes it a Managed Service](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-Introduction/SageMaker-What-Makes-it-a-Managed-Service/page.md): Explains managed vs unmanaged AWS services and how Amazon SageMaker abstracts infrastructure for ML workflows, highlighting benefits, trade-offs, and key features like notebooks, training, and hosting
- [Introduction to Jupyter Notebooks](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-Introduction/Introduction-to-Jupyter-Notebooks/page.md): Overview of Jupyter Notebooks as interactive, web-based tools for exploratory data analysis, visualization, reproducible experiments, collaboration, and comparisons with REPLs and IDEs
- [Introduction to Jupyter Notebooks Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-Introduction/Introduction-to-Jupyter-Notebooks-Part-2/page.md): Overview of running Jupyter notebooks locally, in containers, IDEs, and AWS SageMaker, covering server architecture, instance sizing, examples, visualizations, and workflow benefits
- [Demo Jupyter Notebooks](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-Introduction/Demo-Jupyter-Notebooks/page.md): Guide to install and run Jupyter Notebook and JupyterLab on an Ubuntu EC2 instance, covering virtual environments, connecting remotely, basic usage, and security considerations.
- [The SageMaker SDK for Python Why not boto3](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-Introduction/The-SageMaker-SDK-for-Python-Why-not-boto3/page.md): Explains why the SageMaker Python SDK simplifies ML workflows compared to boto3 by offering higher level abstractions, notebook ergonomics, integrations, and recommends hybrid use for non-ML AWS services

### SageMaker User Interface 

- [Navigating the UI](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/Navigating-the-UI/page.md): Explains navigating the Amazon SageMaker console, differences between Studio and legacy notebooks, and where processing, training, models, endpoints, and notebooks appear.
- [Demo Navigating the UI](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/Demo-Navigating-the-UI/page.md): Guide to navigating the Amazon SageMaker console and creating and using legacy managed notebook instances, locating processing, training, models, endpoints, and using Jupyter and JupyterLab
- [SageMaker Domains and SageMaker Studio](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/SageMaker-Domains-and-SageMaker-Studio/page.md): Overview of Amazon SageMaker Domains and Studio, explaining benefits over notebook instances, architecture, setup options, user profiles, and launching Studio for collaborative ML workflows
- [SageMaker Domains and SageMaker Studio Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/SageMaker-Domains-and-SageMaker-Studio-Part-2/page.md): Guide to SageMaker Studio apps, spaces, EBS-backed storage, private versus shared collaboration, lifecycle and billing, and when to create separate domains for isolation.
- [SageMaker Domains and SageMaker Studio Part 3](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/SageMaker-Domains-and-SageMaker-Studio-Part-3/page.md): Guide to adding SageMaker Studio user profiles to a Domain, configuring apps and storage, distinguishing Studio new EBS from Classic EFS, and following security and resource best practices
- [Options for Collaboration in SageMaker Studio](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/Options-for-Collaboration-in-SageMaker-Studio/page.md): Explains collaboration in SageMaker Studio, comparing EBS/EFS storage limits and advocating using Git inside JupyterLab for versioning, branching, reviews, and reproducible ML workflows.
- [Code Editor Alternative to JupyterLab](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/Code-Editor-Alternative-to-JupyterLab/page.md): Overview of SageMaker Studio Code Editor as a VS Code–based IDE alternative to JupyterLab for structured development, debugging, Git integration, and AWS resource support
- [Code Editor Alternative to JupyterLab Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/Code-Editor-Alternative-to-JupyterLab-Part-2/page.md): Comparison of AWS SageMaker managed VS Code and JupyterLab, guiding when to use each, provisioning Code Editor spaces, debugging, refactoring notebooks into production scripts
- [SageMaker Studio Classic vs SageMaker Studio New](https://notes.kodekloud.com/docs/AWS-SageMaker/SageMaker-User-Interface/SageMaker-Studio-Classic-vs-SageMaker-Studio-New/page.md): Comparison of SageMaker Studio Classic and Studio New, highlighting architectural differences, improved performance, clearer compute model, expanded IDEs, cost control, and migration guidance

### Persona SageMaker Activities - Data Engineer

- [Tabular Data Preparation](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/Tabular-Data-Preparation/page.md): Guidelines and Python examples for cleaning and preparing tabular data for machine learning including missing value handling, categorical encoding, scaling, outlier treatment, and processing environments
- [SageMaker Canvas Low Code Data Preparation and ML Training](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/SageMaker-Canvas-Low-Code-Data-Preparation-and-ML-Training/page.md): Explains using Amazon SageMaker Canvas and Data Wrangler for low-code tabular data preparation, AutoML training, and rapid proof-of-concept model deployment without programming
- [SageMaker Canvas Low Code Data Preparation and ML Training Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/SageMaker-Canvas-Low-Code-Data-Preparation-and-ML-Training-Part-2/page.md): Overview of SageMaker Canvas and Data Wrangler for low-code data preparation, AutoML model building, explainability, deployment, and cost considerations for tabular ML workflows.
- [Demo SageMaker Canvas Data Wrangler](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/Demo-SageMaker-Canvas-Data-Wrangler/page.md): Demonstration of SageMaker Canvas Data Wrangler for no-code data preparation including DQI reporting, dropping and imputing features, scaling and encoding, and exporting ML-ready datasets.
- [Demo SageMaker Canvas Data Wrangler Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/Demo-SageMaker-Canvas-Data-Wrangler-Part-2/page.md): Guide showing how to export SageMaker Data Wrangler transformations to Canvas datasets, Amazon S3, or a Jupyter notebook plus .flow for reproducible Processing jobs, with step examples and tips
- [Demo SageMaker Canvas AutoML](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/Demo-SageMaker-Canvas-AutoML/page.md): Using SageMaker Canvas AutoML with a Data Wrangler processed dataset to build, evaluate, and deploy a tabular regression model including inspection of training artifacts and predictions
- [SageMaker Data Processing Jobs](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Engineer/SageMaker-Data-Processing-Jobs/page.md): Explains using AWS SageMaker processing jobs to offload scalable, reproducible data preprocessing from notebooks, with examples, processor choices, and monitoring guidance.

### Persona SageMaker Activities - Data Scientist 

- [An Introduction to Feature Engineering](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/An-Introduction-to-Feature-Engineering/page.md): Introduction to feature engineering techniques with pandas and scikit-learn, covering encoding, transformations, aggregations, normalization, and SageMaker Processing for reproducible production pipelines
- [An Introduction to Feature Engineering Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/An-Introduction-to-Feature-Engineering-Part-2/page.md): Explains how targeted feature engineering improves model accuracy, generalization, interpretability, and training efficiency, plus strategies for handling data issues and tools for scalable transformations.
- [Demo Feature Engineering in SageMaker Studio](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Demo-Feature-Engineering-in-SageMaker-Studio/page.md): Demonstrates interactive feature engineering in SageMaker Studio with pandas and scikit learn, then offloads batch target encoding to a SageMaker Processing job and stores outputs in S3
- [Training Model in SageMaker Studio and Monitoring Training Jobs](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Training-Model-in-SageMaker-Studio-and-Monitoring-Training-Jobs/page.md): How to run and monitor managed training jobs in Amazon SageMaker Studio, including setup, SDK examples, scaling, cost optimization, logging, checkpoints, and storing artifacts in S3
- [Training Model in SageMaker Studio and Monitoring Training Jobs Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Training-Model-in-SageMaker-Studio-and-Monitoring-Training-Jobs-Part-2/page.md): Explains creating and running SageMaker training jobs with Estimator classes, custom containers, hyperparameters, and automated hyperparameter tuning, plus examples and S3/IAM configuration reminders
- [Training Model in SageMaker Studio and Monitoring Training Jobs Part 3](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Training-Model-in-SageMaker-Studio-and-Monitoring-Training-Jobs-Part-3/page.md): Overview of Amazon SageMaker managed training jobs, Estimator usage, hyperparameter tuning, and compute sizing.
- [Demo Training Your Model in SageMaker Studio Using Python SDK](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Demo-Training-Your-Model-in-SageMaker-Studio-Using-Python-SDK/page.md): Shows how to train and tune a Linear Learner model in Amazon SageMaker Studio using the SageMaker Python SDK Estimator with CSV data uploads and S3 model artifacts
- [Introduction to Experiment Management](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Introduction-to-Experiment-Management/page.md): Guide to using Amazon SageMaker Experiments to track, log, and reproduce machine learning runs using the SDK, Run contexts, and compare results across trials.
- [Optimizing the Training Process with Automation](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Optimizing-the-Training-Process-with-Automation/page.md): How to use Amazon SageMaker Autopilot to automate preprocessing, model selection, hyperparameter tuning, and deployment for faster prototyping and cost controlled iterative tabular model training
- [Keeping Track of Models Using the SageMaker Model Registry](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Keeping-Track-of-Models-Using-the-SageMaker-Model-Registry/page.md): Managing model versions and approvals using Amazon SageMaker Model Registry for organized tracking, deployment automation, governance, and rollback
- [Keeping Track of Models Using the SageMaker Model Registry Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Keeping-Track-of-Models-Using-the-SageMaker-Model-Registry-Part-2/page.md): Describes how data scientists and governance officers use the SageMaker Model Registry to register, review, approve, version, and deploy models with governance and CI/CD examples
- [Demo Keeping Track of Models Using the SageMaker Model Registry](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-Data-Scientist/Demo-Keeping-Track-of-Models-Using-the-SageMaker-Model-Registry/page.md): Shows how to register, version, approve, and deploy machine learning models using Amazon SageMaker Model Registry with the Python SDK and Boto3

### Persona SageMaker Activities - MLOps Engineer

- [Options to Host a Model for Inference](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Options-to-Host-a-Model-for-Inference/page.md): Describes options for hosting ML models for inference and recommends AWS SageMaker Endpoints for managed, low latency, autoscaling production deployments.
- [Options to Host a Model for Inference Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Options-to-Host-a-Model-for-Inference-Part-2/page.md): Comparison of using Boto3 versus the SageMaker Python SDK to create and invoke Amazon SageMaker real time endpoints, highlighting tradeoffs, workflows, and a hybrid approach
- [Advanced Inference Options](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Advanced-Inference-Options/page.md): Explains SageMaker advanced inference options, focusing on Batch Transform for cost-effective offline, periodic, and high-throughput predictions, including workload distribution, transient instances, and mini-batching.
- [Advanced Inference Options Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Advanced-Inference-Options-Part-2/page.md): Explains serverless SageMaker inference for cost-efficient intermittent real-time workloads and using SageMaker Feature Store to precompute and serve consistent low-latency features for inference.
- [Advanced Inference Options Part 3](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Advanced-Inference-Options-Part-3/page.md): Explains Amazon SageMaker inference pipelines for chaining preprocessing, model serving, and postprocessing in a single real-time endpoint, including deployment, comparisons, and best practices.
- [Demo Deploy a Hosted Model Using SageMaker Endpoints](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Demo-Deploy-a-Hosted-Model-Using-SageMaker-Endpoints/page.md): Tutorial for deploying a trained model to Amazon SageMaker by registering model packages, creating endpoint configurations and endpoints, and invoking predictions using Boto3 and the SageMaker Python SDK
- [Demo Deploy a Hosted Model Using SageMaker Endpoints Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Demo-Deploy-a-Hosted-Model-Using-SageMaker-Endpoints-Part-2/page.md): Guide to deploying and invoking Amazon SageMaker real-time endpoints using boto3 sagemaker-runtime and the SageMaker Python SDK, including testing, decoding responses, and cleanup steps.
- [Automate ML Pipeline with SageMaker Pipeline](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Automate-ML-Pipeline-with-SageMaker-Pipeline/page.md): Describes automating ML workflows with Amazon SageMaker Pipelines, using scripts and the Python SDK for reproducible, scalable pipelines integrated with CI/CD and AWS services
- [End to End CICD ML Pipelines from SageMaker Project Templates](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/End-to-End-CICD-ML-Pipelines-from-SageMaker-Project-Templates/page.md): Describes SageMaker project templates that provision CloudFormation CI and CD stacks, link Git repos and CodeBuild to SageMaker Pipelines, automating model build, registry, deployment and governance.
- [Demo End to End CICD ML Pipelines from SageMaker Project Templates](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Demo-End-to-End-CICD-ML-Pipelines-from-SageMaker-Project-Templates/page.md): Demo of end-to-end SageMaker project templates provisioning CI/CD for ML workflows that build, register, approve and deploy models
- [Monitoring and Observability](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Monitoring-and-Observability/page.md): Monitoring and observability for SageMaker models using AWS CloudWatch and SageMaker Model Monitor to track infrastructure health, model drift, data quality, latency, logs, alarms, and remediation
- [Monitoring and Observability Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Monitoring-and-Observability-Part-2/page.md): Guide to monitoring and observability for ML systems using AWS CloudWatch and SageMaker Model Monitor to detect infrastructure issues, model drift, bias, feature attribution changes, and enable automated remediation.
- [Demo Add Monitoring to Deployed Model](https://notes.kodekloud.com/docs/AWS-SageMaker/Persona-SageMaker-Activities-MLOps-Engineer/Demo-Add-Monitoring-to-Deployed-Model/page.md): Demonstrates adding SageMaker Model Monitor data quality monitoring to a deployed endpoint by creating a baseline, enabling data capture, scheduling monitoring, generating inference traffic, and inspecting results.

### Wrap Up

- [Demystifying Advanced SageMaker](https://notes.kodekloud.com/docs/AWS-SageMaker/Wrap-Up/Demystifying-Advanced-SageMaker/page.md): Overview of SageMaker advanced features, comparing foundation models, Bedrock, and HyperPod clusters with guidance on fine‑tuning, hosting, scaling, and decision criteria.
- [Demystifying Advanced SageMaker Part 2](https://notes.kodekloud.com/docs/AWS-SageMaker/Wrap-Up/Demystifying-Advanced-SageMaker-Part-2/page.md): Advanced SageMaker practices for production ML including human-in-the-loop inference, scalable data labeling with Ground Truth, and managed RStudio integration with Python interoperability
- [Demystifying Advanced SageMaker Part 3](https://notes.kodekloud.com/docs/AWS-SageMaker/Wrap-Up/Demystifying-Advanced-SageMaker-Part-3/page.md): Explains using MLflow with Amazon SageMaker to unify experiment tracking, model registry, and multi-platform deployment for consistent ML lifecycle management
- [Whats New for 2025](https://notes.kodekloud.com/docs/AWS-SageMaker/Wrap-Up/Whats-New-for-2025/page.md): Overview of Amazon SageMaker changes in 2025, detailing the split into SageMaker AI and SageMaker platform, Unified Studio, Projects, Spaces, governance, lakehouse and analytics integration
- [Is SageMaker Demystified for You](https://notes.kodekloud.com/docs/AWS-SageMaker/Wrap-Up/Is-SageMaker-Demystified-for-You/page.md): Overview of AWS SageMaker course covering end-to-end ML pipeline, tools, deployment, and next steps

### Course Introduction 

- [Course Overview](https://notes.kodekloud.com/docs/AWS-SageMaker/Course-Introduction/Course-Overview/page.md): A practical hands on AWS SageMaker course teaching end to end ML pipelines, labs, and role based workflows for data engineers, data scientists, and MLOps engineers.