AI Engineering Foundations

6 months duration

8 modules

Updated Jul 30, 2026

Development & Programming

Course Overview

Get to know what this course is all about and what you'll learn

Course Description

AI Engineering Foundations is a seven-week, build-first program. It begins with Python from the ground up — so no coding experience is needed — then moves quickly into working with large language models: prompting, retrieval, and simple agents. You finish by shipping a working AI application.

What you'll learn

Write Python using core syntax, data structures, and functions
Work with web APIs and JSON from Python
Call large language model APIs and craft effective prompts
Build applications with an LLM app framework
Implement retrieval-augmented generation (RAG) over your own data
Assemble a simple AI agent and evaluate its output
Build and demo a working AI-powered application

How it's delivered

Weekday lessons are on-demand (Monday–Thursday); live weekend sessions apply each week's topic with a mentor. The course ends with a capstone project. Submit every exercise and your capstone through your workspace.

What You'll Learn

Build real AI-powered apps in seven weeks. AI Engineering Foundations starts with Python from scratch, then takes you through LLM APIs, prompt engineering, retrieval-augmented generation, and agents — ending with an app you build and demo.

Course Curriculum

8 modules • Instructor-led • Hands-on projects

Course Curriculum

Prerequisites

Comfort using a computer and logical thinking. No prior coding required — Python is taught from the start. A laptop and an internet connection; access to AI APIs is provided or guided. (Classical ML, MLOps, and fine-tuning are covered in the full program.)

Learning Objectives

Write Python using core syntax, data structures, and functions
Work with web APIs and JSON from Python
Call large language model APIs and craft effective prompts
Build applications with an LLM app framework
Implement retrieval-augmented generation (RAG) over your own data
Assemble a simple AI agent and evaluate its output
Build and demo a working AI-powered application

Course Modules

Master the fundamental tool that every professional developer uses daily. Learn to track changes, collaborate with others, and manage your code like a pro from the very beginning of your development journey.

What you'll learn

Understand version control concepts and why Git is essential for modern software development
Use GitHub effectively for remote repositories, collaboration, and showcasing your work to potential employers
Master Git basics including repositories, commits, branches, and merging for effective code management.

Python serves as the foundation of modern data science, providing essential programming skills for data manipulation, analysis, and machine learning. This module develops your Python proficiency from basics through data science applications.

You'll master Python fundamentals including data types, control structures, functions, and essential libraries. Hands-on exercises with real datasets teach you to write efficient code for data processing tasks and establish the foundation for advanced data science work.

What you'll learn

Write Python programs using core data types, variables, and operators
Control program flow with conditionals, loops, and comprehensions
Organise code into reusable functions and modules
Work with lists, dictionaries, sets, and tuples to structure data
Read from and write to files, handling common errors gracefully
Apply these fundamentals to simple data-manipulation tasks

Use Python to talk to the web. Work fluently with core data structures and JSON, then call REST APIs with the requests library and handle their responses.

What you'll learn

Work confidently with Python lists, dicts, and JSON
Make HTTP requests with the requests library
Parse and navigate JSON API responses
Handle errors, status codes, and auth headers

Move from "asking an LLM nicely" to a discipline — structured prompts, chain-of-thought, output formatting, prompt injection defence, testing, and versioning. The cheapest, most underrated lever in any AI system.

What you'll learn

Principles of effective prompting
Prompt structure: instruction, context, input, output format
Zero-shot prompting
Few-shot prompting and example selection
Chain-of-thought (CoT) prompting
Self-consistency prompting
Tree of thoughts
Role and persona prompting
System prompts design and best practices
Output formatting: JSON mode, structured outputs, delimiters
Prompt chaining and decomposition
Prompt templates and reusability
Handling edge cases and failures
Prompt injection and security considerations
Prompt testing and evaluation
Prompt versioning and management
Domain-specific prompting strategies

Stand on the shoulders of LangChain, LlamaIndex, Haystack, and MCP. Patterns for composing models, retrievers, memory, and tools without reinventing the plumbing every project.

What you'll learn

LangChain: models, prompts, output parsers
LangChain Expression Language (LCEL)
Memory systems: buffer, window, summary
Callbacks and tracing with LangSmith
Document loaders and text splitters
Retrievers and retrieval chains
Tools and function calling
Structured outputs with Pydantic
LlamaIndex fundamentals
Haystack overview
Model Context Protocol (MCP) for tool integration
Design patterns for AI applications
Framework-agnostic development

Give an LLM access to your data without fine-tuning. Embedding models, vector databases, chunking strategies, hybrid search, re-ranking, advanced patterns (parent-document, RAPTOR, Graph RAG), evaluation, and production deployment.

What you'll learn

RAG architecture and components
Embedding models: OpenAI, Cohere, BGE, E5
Vector databases: Pinecone, Weaviate, Qdrant, Chroma, pgvector
Document processing and chunking strategies
Query rewriting and multi-query approaches
Hybrid search (keyword + semantic)
Re-ranking with cross-encoders
Advanced RAG: parent document, sentence window, RAPTOR, Graph RAG
Citation and source attribution
RAG evaluation with RAGAS
Production RAG: scaling, incremental indexing, access control

From single-tool ReAct loops to multi-agent crews. LangGraph state machines, tool design, memory, safety guardrails, human-in-the-loop checkpoints, and the observability you need to debug an agent in production.

What you'll learn

Agent architectures: ReAct, Plan-and-Execute, Tree of Thoughts
Tool design and implementation
Agent memory: short-term, long-term, episodic
LangChain agents and AgentExecutor
LangGraph: state management, conditional edges, cycles, checkpointing
Multi-agent systems: CrewAI, AutoGen
Agent evaluation and testing
Safety and guardrails: input validation, action constraints, sandboxing
Human-in-the-loop patterns
Agent observability and debugging

Capstone: a RAG or simple agent application — module in the AI Engineering Foundations course.

What you'll learn

Design a retrieval-augmented or agent-based application end to end
Build a document pipeline with embeddings and a vector store
Ground an LLM's answers in retrieved context to reduce hallucination
Orchestrate tools or multi-step reasoning where the task needs it
Evaluate the application's accuracy, cost, and latency
Ship a working demo and walk through its architecture