Start with the learning job
If AI still feels like a collection of product names, begin with concepts. If you already understand the basic vocabulary and your problem is getting more reliable output from a language model, move toward prompt engineering. The order matters because prompting techniques make more sense when you understand what the model can and cannot do.
Step 1: build the conceptual map
Master Generative AI is the lighter orientation step. Use it to learn the vocabulary, then write down which concepts are still unclear. Do not try to memorise every model or product name; those change faster than the underlying ideas.
Step 2: practice prompting
The ChatGPT and AI Masterclass is a longer next step focused more directly on prompt engineering. Treat every prompting technique as a hypothesis to test on your own work rather than as a universal formula.
Step 3: create a real exercise
Choose one repeated task: summarising research, restructuring notes, drafting code explanations or comparing alternatives. Build a small test set and compare prompts against the same task. That turns course content into observable skill.
What not to expect
Neither course makes model output automatically true. AI work still needs source checking, domain judgment and awareness of privacy or confidential-data constraints.
Decision rule
Take the shorter conceptual course when vocabulary is the bottleneck. Take the longer prompt-engineering course when you already know why you want to use AI and need a more deliberate interaction method.