Ensure every participant can confidently build and deploy production-grade applications on AWS. Covers AWS architecture, security, serverless patterns, Infrastructure as Code, and end-to-end CI/CD pipelines. Week 3 culminates in a fully deployed, monitored cloud application.

GenAI systems need reliable, scalable data pipelines. These two weeks cover modern data lake architecture on AWS, real-time streaming, ETL orchestration, and feature pipelines — the data infrastructure that feeds ML and LLM systems.

A production-focused introduction to machine learning for experienced engineers. Not a statistics course - this week covers the ML lifecycle as it actually runs in enterprise environments, using SageMaker and MLOps tooling.

Understand how large language models work in practice and how to get reliable, high-quality output from them. Covers transformer architecture at a practical level, prompt engineering patterns, and evaluation techniques essential for production LLM systems.

Retrieval-Augmented Generation (RAG) is the most widely deployed GenAI pattern in enterprise production. This week covers the full RAG stack from embeddings and vector databases through to retrieval evaluation — and culminates in building a production-grade RAG chatbot.

Deploy foundation models and build autonomous AI agents using Amazon Bedrock. Covers Bedrock's model library, guardrails for safe deployment, and agent architectures using LangChain and LangGraph for real-world autonomous workflows.

Move from building LLM systems to running them at scale in production. Covers fine-tuning, observability, evaluation frameworks, multi-agent orchestration, and the operational concerns of scaling GenAI — the skills that separate a prototype from a production system.

Apply everything learned across the program by building a complete, deployable production GenAI system. Choose one of three specialization tracks. The capstone is your primary portfolio piece and the centerpiece of your technical interview preparation.