Generative AI and Agents / 7 weeks
Generative AI and Large Language Models
Teaches operators how to read what generative models actually do, where they fail, how to ground them in real data, and how to evaluate output quality before it reaches a customer.
Learning path
Six modules with practical application.
- What LLMs and diffusion models actually do, where their fluency is real, and where fluent confidence is the most dangerous signal
- Zero shot, few shot, chain of thought, role prompting, and structured output prompts that get LLMs to do useful work reliably
- Grounding LLM outputs in your own documents and data so answers are current, specific, and citeable
- Defining quality for generative output and measuring it with rubrics, paired comparisons, and reference sets that do not flatter
- Filtering inputs, controlling outputs, blocking jailbreaks, and stopping prompt injection before it reaches your data
- Picking the right model, sizing cost, latency, and reliability, and writing a ship memo that survives operating reality
Loading course material…