Deep Learning on AWS
From Developer to Full Stack Deep Learning Engineer on AWS. This curriculum trains you to become a Full Stack Deep Learning Engineer, capable of not just training models but also deploying and managing them in production for business value. You will learn to build, step-by-step, production-grade Deep Learning microservices in AWS. By the end, you’ll build and publish a capstone project on GitHub showcasing your skills in managing AI through its entire lifecycle.
Who is this for?
Developers who want to build, deploy, and manage production-grade Deep Learning models on AWS.
Prerequisites:
You must bring an AWS account (a free one is sufficient) to run the hands-on labs. We assume you have some coding experience in some language (examples are primarily in Python).
What You Will Achieve
- Build, test, and deploy production-grade Image Classifiers and a custom project (image or time-series).
- Master the full production AI lifecycle in AWS, from data ingestion in S3 to application integration.
- Gain deep, hands-on experience with AWS Sagemaker, Lambda, API Gateway, S3, and GPU instances.
- Learn to hyper-parameter tune production-grade neural networks like ResNet (for images) and DeepAR (for time-series).
- Understand and implement the microservice design pattern for AI deployment.
Key Topics Covered
This 8-session, 1.5-hour-per-session curriculum is structured around in-demand business use cases:
- Production AI Cloud Tools: AWS Sagemaker (Built-in Algorithms and Transfer Learning), Sagemaker Endpoint, AWS Lambda, AWS API Gateway, AWS Roles, AWS Cloudwatch, and AWS S3.
- AI Algorithms & Concepts: Production-grade Deep Learning (DL), CNN (ResNet) for images, DeepAR for time-series forecasting, Transfer Learning, and Recurrent Neural Networks (RNNs).
- Lifecycle Management: Bringing data into Sagemaker, Model Training and Validation, Hyper-parameter tuning, Python-based API testing, and an intro to model versioning, diagnosis, and retraining.
Assessment & Certification
ssessment is based on building and testing Image Classifiers and completing a capstone project where you will build and publish your own custom AI application on GitHub. This includes gathering data, training a neural network, building a prediction service, and integrating it into an application.