Master AI from <span class="text-gradient inline-block" style="opacity:0;transform:translateY(0.7em)">first principles.

The AI Shift

How modern AI differs from traditional software.

~12 min Quiz

LLM Mechanics

How language models actually predict text.

~12 min Quiz

The Boundaries

What today's AI can and can't do.

~11 min Quiz

Lightweight Corporate Guardrails

Simple rules for safe AI use at work.

~10 min Quiz

Project

AI in your world

Map where AI fits in your own role.

~12 min

Start project

Level 2Foundations

Applied Prompt Engineering & Interaction

Ask sharper questions, get dramatically better answers.

6 lessons + project

Prompt Mechanics

How prompts are read, token by token.

Core Prompting

The building blocks of an effective prompt.

Advanced Frameworks (CoT, few-shot, persona)

Chain-of-thought, few-shot, and persona techniques.

Context Constraints

Working within the model's context window.

Staying Safe: Data Leakage & Prompt-Injection Basics

Avoid data leakage and prompt injection.

Trust but Verify: Fact-Checking AI Output

Fact-check AI output before you trust it.

TBD

Build and test a reusable prompt.

Level 3Adoption & Models

The AI Economy & Workplace

Spot the real opportunities - and the true cost of AI at work.

4 lessons + project

AI Ergonomics

Fit AI smoothly into your daily work.

Core GenAI Economics

What generative AI really costs to run.

Use-Case Discovery

Find high-value places to apply AI.

AI Change Management (Shadow AI)

Roll out AI and tame shadow usage.

TBD

Pitch an AI use case with its ROI.

Level 4Adoption & Models

Next-Gen Model Paradigms

Reasoning, multimodal, open models: know which to reach for.

4 lessons + project

Inference-Time Compute (reasoning models)

How reasoning models think before answering.

Multimodality

Working with image, audio, and text together.

Model Flavours (OSS / API / SLM)

Choosing between open, API, and small models.

System Strategy (build vs buy, vendor eval)

Build vs buy, and evaluating vendors.

TBD

Recommend the right model for a use case.

Level 5Adoption & Models

AI Automation Workflows

Wire AI into workflows that save hours every week.

6 lessons + project

Workflow Design & Architecture

Design automations that actually hold up.

No-Code Automation (Zapier, Make)

Automate work with Zapier and Make.

API & Webhook Connections

Connect tools using APIs and webhooks.

Human-in-the-Loop Intercepts

Add human checkpoints to AI workflows.

Computer Use & Desktop OS Agents

Agents that operate your desktop apps.

AI Workflow ROI

Measure the hours your workflows save.

TBD

Ship an automation that saves real time.

Level 6Quality & Engineering

AI Evaluation & Reliability

Prove your AI works - and keep it working in production.

5 lessons + project

Benchmark Basics

What benchmarks do and don't tell you.

AI Failure Modes

Recognise how and why AI breaks.

Core Measurement Metrics

The metrics that measure AI quality.

LLM-as-a-Judge

Use a model to grade model output.

Continuous / Production Evaluation & Drift Monitoring

Catch drift with ongoing evaluation.

TBD

Build an eval suite for an AI feature.

Level 7Quality & Engineering

Spec-Driven Dev & Deep Economics

Steer AI with specs and tools, and master the economics.

4 lessons + project

Spec-Driven Development (AGENTS.md / rules files)

Steer AI with specs and rules files.

Tool Use Fundamentals

Let models call tools and functions.

Model Context Protocol (MCP)

Connect AI to tools via MCP.

Deep GenAI Economics

Optimise cost, latency, and throughput.

TBD

Wire a model up to real tools.

Level 8Data & RAG

Data Foundations & Basic RAG

Give AI your own knowledge so it stops guessing.

4 lessons + project

Vector Spaces (embeddings)

How embeddings turn meaning into numbers.

RAG Architecture

Ground AI answers in your own documents.

Data Preparation (chunking)

Chunk and clean data for retrieval.

Data Governance

Keep retrieval data secure and compliant.

TBD

Build a basic RAG over your docs.

Level 9Data & RAG

Advanced Retrieval & Search Optimisation

Make retrieval fast, precise, and trustworthy at scale.

4 lessons + project

Retrieval Optimisation

Make retrieval faster and more precise.

Advanced Search Patterns

Hybrid and re-ranked search strategies.

Advanced RAG Tactics (Multi-query, HyDE, PRF; define/replace 'DSD')

Multi-query, HyDE, and feedback retrieval.

RAG Evaluation (BERTScore, ROUGE, faithfulness)

Measure retrieval relevance and faithfulness.

TBD

Tune a RAG pipeline for accuracy.

Level 10Agents & Governance

Autonomous Single-Agent Systems

Build agents that plan, use tools, and finish the job.

6 lessons + project

Agent Design & Logic (ReAct)

Design agents that reason, then act.

Agent Memory (short-term vs episodic long-term)

Give agents short- and long-term memory.

Stateful Graph Orchestration (LangGraph)

Orchestrate agent state with LangGraph.

Engineering Lifecycle (observability, tracing)

Trace and observe agents in production.

Execution Controls & Structural Guardrails

Keep agents inside safe boundaries.

Agent Evaluation

Test whether your agent actually works.

TBD

Build an agent that finishes a task.

Level 11Agents & Governance

Multi-Agent Swarms & Corporate Governance

Coordinate many agents safely - and govern the risk.

5 lessons + project

Multi-Agent Systems (supervisor/worker, Agent + RAG)

Coordinate supervisor and worker agents.

AI Apps Workflows (enterprise examples)

Patterns from real enterprise AI apps.

Agent Governance & Liability

Who's accountable when agents act.

Deep Corporate Guardrails

Enterprise-grade controls for AI systems.

LLM Security & the OWASP LLM Top 10

Defend against the top LLM threats.

TBD

Design a governed multi-agent system.

Level 12Deep Architecture

Model Adaptation & Deep Architecture

Look under the hood: transformers, tuning, and infrastructure.

4 lessons + project

Deep Architectures (Transformers, Attention)

Inside transformers and the attention mechanism.

Alternative Paradigms (Diffusion, World Models)

Diffusion and world models, explained.

Efficiency, Tuning & Accuracy (Quantisation, LoRA)

Quantisation and LoRA fine-tuning.

Infrastructure Operations (GPU, edge SLM, on-prem)

Run models on GPU, edge, and on-prem.