Generative AI Study Plan
Build from a bounded prompt to a tested application boundary.
How it works
Begin with one supplied-source transformation and a human acceptance guide. Learn context, output formats and iteration before comparing tools. Next practise source checking, prompt injection awareness and multimodal input limits. Finish with validators, tool policies, evaluation cases and the workshop project. Keep prompt drafts, reference cases and actual provider observations separate so a later reviewer can reproduce what was really tested.
How to use this lesson
This is an authored prompt exercise with a human-written acceptance guide. It is not a captured provider response or a promise that a model will produce the target. No provider account, API, model weights or live inference was used for the course execution review. If you try it in an external service, use an authorised environment, check its actual access and data settings, and record the product/model and run date.
Bhylu does not run a generative model or send this prompt to an AI service. Copy and download preserve the text only. External tools may require accounts, paid access or compatible inputs; this course does not promise availability or prices.
Prompt exercise
Create a personal practice checklist with four stages:
1. Supplied-source prompt.
2. Missing-evidence and adversarial cases.
3. Local validation and permission labs.
4. A separately authorised provider evaluation.
Do not mark a stage complete without its evidence.
Acceptance guide — not a model transcript
The target checklist keeps local-code evidence separate from actual provider-run evidence.
Reading the course does not provide certification or prove production readiness. Revisit any stage where you cannot explain the failure behaviour. If a service, model or account setting changes, rerun the affected real evaluation instead of relying on a historical screenshot or an unchanged prompt filename.