Lesson 1
From deterministic software to probabilistic systems
Traditional software follows explicit logic. Generative AI predicts outputs from context, so correctness must be evaluated differently.
All learning tracks
Applied AI EngineeringFoundation 60–90 min
AI Foundations for Engineers
Understand models, prompts, tokens, context, temperature, structured outputs and where AI systems fail.
LLMsPromptsContext
Learning outcomes
Explain what an LLM is and is not
Recognise common failure modes
Choose deterministic vs generative behaviour
Guided lessons
Learn the engineering thinking
Lesson 2
Prompt, context and model responsibilities
Separate instructions, business context, retrieved evidence and model behaviour so failures can be diagnosed rather than guessed.
Lesson 3
Temperature and output variability
Learn when variability is useful and when business workflows need constrained, structured outputs.
Lesson 4
Structured outputs and schemas
Use predictable data contracts for downstream automation instead of parsing free-form text.
Lesson 5
Failure modes engineers should expect
Hallucination, omission, overconfidence, stale knowledge, ambiguous instructions and unsafe tool use must be designed for.
Hands-on practice
Do something with what you learned
Do not just tick these mentally. Write the query, create the test matrix, refactor the code, or document the decision. Practical evidence is the point.