Complete Syllabus

Your path to
AI engineer in 12 weeks

  • 9modules
  • 110topics
  • 5projects
  • 120hours
Generative AI Masterclass 9 modules · 5 projects · 120 hours
  • 01Generative AI Foundations 12 topics · ~1 week · 10 hrs
    • Core concepts
      • AI, machine learning and deep learning explained
      • What makes AI “generative”
      • How generative AI evolved: a timeline
      • The GenAI landscape: models, vendors and tools
    • Working with LLMs
      • Introduction to large language models
      • Architecture of an LLM application
      • Managing conversation history
      • Tokens and how pricing works
      • Context windows and their limits
    • Quality and behaviour
      • Prompt engineering vs context engineering
      • Understanding model behaviour
      • Evaluating outputs with human review
  • 02AWS Cloud for AI Deployment 15 topics · 1 project · ~1.5 weeks · 16 hrs
    • Getting started
      • Creating and securing your AWS account
    • Identity and access (IAM)
      • Users without console access
      • Assigning permissions
      • Working with groups
      • Service accounts and the AWS CLI
      • Roles
    • Compute, storage and databases
      • Launching and connecting to EC2
      • S3 buckets: storing and accessing data
      • RDS: creating and connecting to a database
    • Containers and serverless
      • Running a Python app on EC2
      • Docker fundamentals
      • Pushing images to Amazon ECR
      • AWS Lambda functions
      • Lambda with container images
      • Publishing APIs with API Gateway
    • ProjectServerless REST APILambda functions backed by DynamoDB, published to the web through API Gateway.
  • 03Python for AI Engineering 25 topics · 3 projects · ~2.5 weeks · 24 hrs
    • Setup
      • Installing and configuring VS Code
      • Where Python fits: domains and platforms
    • Language basics
      • Syntax and semantics
      • Variables and basic data types
      • Operators
      • Conditions
      • Loops
    • Data structures
      • Lists
      • Sets
      • Tuples
      • Dictionaries
    • Functions and modules
      • Defining and using functions
      • Lambda, map and filter
      • Modules, packages and imports
      • A tour of the standard library
      • Working with files
    • Object-oriented Python
      • Classes and objects
      • Inheritance
      • Polymorphism
      • Encapsulation
      • Abstraction
    • Writing robust code
      • Exception handling
      • Custom exceptions
      • Logging
      • Data validation with Pydantic
    • ProjectMini projectsTwo guided builds: a functions-based utility and an object-oriented mini app.
    • ProjectBanking SystemA console banking app that puts classes, exceptions and logging to work.
    • ProjectPython + AWS Microservices (Capstone)Customers and orders APIs on Lambda and API Gateway with a PostgreSQL database, run locally with Docker Compose.
  • 04LLMs & Prompt Engineering 15 topics · 1 project · ~1.5 weeks · 14 hrs
    • How LLMs work
      • What a large language model is
      • Common LLM problems and limitations
      • Enterprise LLM architecture
      • How LLMs generate text
      • Byte-pair encoding (tokenisation), hands-on
    • Prompt engineering
      • Prompt engineering fundamentals
      • Types of prompting techniques
      • Worked examples
      • Iterative prompt development
      • Summarising
      • Inferring
      • Transforming
      • Expanding
      • Model parameters: temperature, top-p and more
    • Setup
      • Creating OpenAI and Tavily API accounts
    • ProjectText-to-SQL ChatbotAsk questions in plain English; the LLM writes the SQL, runs it on PostgreSQL and replies in a chat assistant.
  • 05RAG & Vector Databases 18 topics · ~1.5 weeks · 16 hrs
    • RAG fundamentals
      • The problem RAG solves
      • How RAG works
      • Traditional vs vector databases
      • Ingestion
      • Retrieval
      • Augmentation and generation
      • End-to-end RAG architecture
      • Estimating vector storage from your files
    • Vector databases in practice
      • Types of vector databases
      • Working with Pinecone
      • Using Hugging Face datasets
      • Building a simple RAG with Pinecone
      • Tips and recommendations
    • Advanced retrieval
      • Two-stage RAG: retrieve, then re-rank
      • Why chunking matters
      • Chunking strategies
      • Context-aware chunking
      • Hybrid search and optimisation
  • 06LangChain Framework 4 topics · ~1 week · 8 hrs
    • LangChain essentials
      • Why LangChain
      • Working with LangChain
    • Advanced patterns
      • Agentic RAG with hybrid search
      • Query translation techniques
  • 07Graph Databases & GraphRAG 9 topics · 1 project · ~1 week · 10 hrs
    • Graph concepts
      • Why graph databases
      • What Neo4j and Cypher are
      • Core components of Neo4j
      • Graphs are everywhere: real-world use cases
    • Cypher hands-on
      • Lab setup
      • Reading data
      • Finding relationships
      • Filtering queries
    • LangChain and graphs
      • Querying a graph database from LangChain
    • ProjectAirline Knowledge GraphGenerate airline data, load it into Neo4j and query it in plain English through LangChain.
  • 08Agentic AI with LangGraph 9 topics · ~1.5 weeks · 14 hrs
    • Foundations
      • Why LangGraph
      • Building a simple graph
      • Debugging with LangSmith Studio
    • Building agents
      • Chains
      • Routers
      • Agents
      • Agents with memory
    • Production readiness
      • Persistence
      • Persistence with PostgreSQL
  • 09Weekend Labs: GenAI on Kubernetes 3 topics · Weekends · 8 hrs
    • Amazon EKS
      • Kubernetes core concepts
      • EKS architecture for GenAI apps
      • Deploying a GenAI workload to EKS