Introduction to AI on AWS
Build a Portfolio of 7 Production AI Microservices on AWS. This curriculum trains you to become a Full Stack Data Scientist, capable of not just training models but also deploying and managing them in production for business value. You will learn your Full Stack DS skills by building 7 production AI microservices in AWS. This course teaches you how to manage an AI’s lifecycle, select the right algorithms for different business use cases, and even build custom algorithms in Docker containers.
Who is this for?
Developers who want to become Full Stack Data Scientists by building a portfolio of real-world, production-grade AI services on AWS.
Prerequisites:
You must bring an AWS account (a free one is sufficient). We assume you have some coding experience in some language (examples are primarily in Python).
What You Will Achieve
- Build and deploy 7 working, production-grade AI services for your professional portfolio.
- Master production AI tools like AWS Sagemaker, Lambda, API Gateway, and S3.
- Learn to bring custom AI algorithms (e.g., for Sentiment Analysis) into AWS Sagemaker as a Docker container.
- Address critical production issues like AI Bias, Data Skew, and Drift.
- Map business problems (like Churn, Pricing, Recommendations) to the right AI algorithms (XGBoost, KNN, LinearLearner).
Key Topics Covered
This 8-session, 1.5-hour-per-session curriculum is structured around in-demand business use cases:
- Business Use Cases: Churn prediction, Pricing Analysis, Customer Approvals, Appointment planning, Sentiment Analysis, and Making Recommendations.
- Production AI Cloud Tools: Sagemaker (Built-in Algorithms and Custom Docker Containers), Sagemaker Endpoint, Lambda, API Gateway, Cloudwatch, S3, Postman/cURL testing.
- AI Algorithms & Concepts: Powerful algorithms like XGBoost, LinearLearner, and KNN; Feature Engineering (One Hot Encoding, Missing Value); and Text Classification (Bag of Words).
- AI Trust & MLOps: AI Trust, Fairness, and Bias; Detecting and removing Skew; Managing AI risk; and an intro to Drift, live monitoring, diagnosis, and model versioning.
Assessment & Certification
Expertise is validated by successfully completing the 7 labs to build 7 working production AI services. Students who complete the course will have all artifacts (models, datasets) in their own AWS account for further use and to showcase to employers.