From using AI to engineering it.

1. Week 1

   Understanding Modern AI

   • AI, machine learning and deep learning
   • Generative AI and foundation models
   • How LLMs work, conceptually: tokens, context, inference
   • Why hallucinations happen, and knowledge cut-offs

   Practical Explore the same tasks across several modern AI systems and start seeing the differences between models, interfaces and capabilities.

2. Week 2

   Working Effectively with AI

   • Instructions, context and constraints
   • Zero-shot and few-shot prompting, structured outputs
   • Working with documents, images, tables and spreadsheets
   • Choosing the right tool, without becoming dependent on one

   Practical Take a difficult real-world task and develop a repeatable AI-assisted workflow for it.

3. Week 3

   Applying AI to Real Work

   • What makes a good AI use case
   • Augmentation versus automation, and human-in-the-loop
   • When conventional software is the better answer
   • From single prompts to repeatable workflows

   Practical Identify a real process from your work, studies or business and redesign it using AI.

4. Week 4

   Responsible AI and Final Project

   • Hallucination, bias, privacy and prompt injection
   • Copyright, confidentiality and data protection
   • Why fluent answers are not necessarily correct
   • Designing human review, and knowing when not to use AI

   Practical Design an AI-assisted solution to a real problem, covering the workflow, where human oversight belongs, the risks and how success is measured.

I understand AI and can use it effectively and responsibly.

1. Week 1

   Python and APIs for AI

   • A focused Python refresher for the whole programme
   • HTTP, REST, authentication, rate limits and error handling
   • Colab, local environments, Git and GitHub

   Practical Build a Python application that sends requests to an AI model and processes the output.

2. Week 2

   Building with Language Models

   • System and user instructions, tokens and context
   • Temperature and sampling, structured outputs, streaming
   • Hosted APIs, open-weight models and Hugging Face

   Practical Build a structured information extraction or classification application.

3. Week 3

   Embeddings and Semantic Search

   • What embeddings represent, and similarity metrics
   • Sentence Transformers, FAISS and nearest-neighbour search
   • Building retrieval directly before reaching for a framework

   Practical Build a semantic search engine over a real collection of documents.

4. Week 4

   Retrieval-Augmented Generation

   • Parsing, chunking, embedding, retrieval and context construction
   • Citations, retrieval failures and hallucination
   • Hybrid retrieval, reranking, knowledge graphs and GraphRAG

   Practical Build a complete RAG application over a real knowledge base.

5. Week 5

   Tools, Function Calling and Agents

   • Tool and function calling, state, memory and agent loops
   • Building an agent loop yourself, before any framework
   • LangGraph, stateful workflows and common failure modes

   Practical Build an AI system that can select and use external tools.

6. Week 6

   Evaluating AI Systems

   • Golden datasets, deterministic and semantic evaluation
   • LLM-as-a-judge; separating retrieval quality from generation quality
   • Prompt injection, data leakage, guardrails and auditability

   Practical Deliberately break the RAG or agent system you built, then improve it from the results.

7. Week 7

   Turning AI into an Application

   • Separating model logic from interfaces; FastAPI
   • Secrets, logging and error handling
   • Gradio interfaces, containers and hosting

   Practical Turn one of your systems into an application someone else can actually use.

8. Week 8

   Capstone Project

   • Architecture, model choice, data and retrieval strategy
   • Evaluation, limitations, risks and deployment approach
   • Published to GitHub with clear documentation

   Practical Design and build an end-to-end AI application you can demonstrate and explain.

I can build and evaluate real AI applications.

1. Week 1

   Deep Learning from First Principles

   • Tensors, parameters, activations and forward propagation
   • Loss functions, gradient descent and backpropagation
   • PyTorch: autograd, modules, optimisers and training loops

   Practical Build and train a neural network in PyTorch and inspect how its parameters change during optimisation.

2. Week 2

   Transformers from First Principles

   • Tokenisation, embeddings and positional information
   • Query, Key and Value projections; scaled dot-product attention
   • Multi-head attention, residuals, normalisation and stacking

   Practical Implement core components of self-attention and inspect the resulting tensor operations.

3. Week 3

   Understanding LLM Architecture

   • Text to tokens to embeddings to logits to next token
   • Hidden dimensions, layers, attention heads, MHA, MQA and GQA
   • Inspecting a real small open-weight model

   Practical Load an open model and trace data through the key components of its architecture.

4. Week 4

   LLM Inference

   • Prefill, decode, autoregressive generation and sampling
   • Compute, memory bandwidth and arithmetic intensity
   • Latency, throughput, time to first token, inter-token latency

   Practical Build a simplified generation loop and benchmark model behaviour.

5. Week 5

   Fine-Tuning and Model Adaptation

   • Prompting versus RAG versus fine-tuning
   • Dataset preparation and training objectives
   • LoRA, QLoRA, PEFT and the training trade-offs

   Practical Prepare a small dataset and adapt an open-weight model using parameter-efficient techniques.

6. Week 6

   Quantisation

   • Precision, GPU memory, bandwidth, speed and model quality
   • FP32, FP16, BF16, INT8 and INT4
   • Scaling, zero points, calibration, GPTQ and AWQ

   Practical Implement a simple quantiser from first principles, then apply existing tooling to an open model.

7. Week 7

   KV Caching and LLM Memory

   • Why previously computed Keys and Values can be reused
   • Cache dimensions, sequence length, batch size and KV heads
   • Fragmentation, paging and why PagedAttention changed serving

   Practical Calculate KV cache requirements for different model configurations and inspect caching during generation.

8. Week 8

   Batching, Scheduling and LLM Serving

   • Static, dynamic and continuous batching
   • Prefill versus decode workloads, queues and admission control
   • Chunked prefill and the architecture of engines such as vLLM

   Practical Run and benchmark an inference workload, and observe how batching changes latency and throughput.

9. Week 9

   Production Deployment and Scaling

   • Model servers, containers, GPU deployment and observability
   • Benchmarking: TTFT, tokens per second, GPU and memory utilisation
   • Data, tensor, pipeline, context and expert parallelism

   Practical Deploy and benchmark a model-serving endpoint.

10. Week 10

    AI Engineering Capstone

    • Model selection, precision, quantisation and hardware requirements
    • KV cache budget, serving configuration and batching
    • Latency, throughput, observability, deployment and cost

    Practical Take an open-weight model and design a production serving solution for a defined workload, then explain the engineering decisions behind it.

I understand how modern AI models work and can optimise, serve and deploy them.