Gen AI + Prompt Engineering

Learning Outcomes

✓ Explain how LLMs work at a high level and key limitations

✓ Apply prompt patterns (instruction, few-shot, chain-of-thought, style/role, tool-use)

✓ Evaluate and iterate prompts with objective rubrics and lightweight metrics

✓ Build RAG with a simple retriever and guardrails; extend to multimodal I/O

✓ Orchestrate small agent workflows and integrate via REST APIs

Curriculum

12-Week Curriculum

Week	Session A	Session B	Micro-lab
1	GenAI landscape & use-cases	LLM fundamentals: tokens, context, safety	Prompt a public model; note latency & cost
2	Prompt engineering basics	Advanced prompting (few-shot, style, structure)	Rewrite prompts using patterns; compare outputs
3	Prompt optimization & evaluation	Chatbot basics with system/user/assistant roles	Build a rubric; A/B test two prompts
4	Text generation & automation	Code generation & analysis	Write a small script generated by an LLM
5	Image generation I (concepts, prompts)	Image generation II (parameters, safety)	Create an image prompt book (3 variants)
6	Practice set: real-world tasks	Prompt engineering deep-dive I (optimization)	Submit an improved prompt + evidence
7	Prompt engineering deep-dive II (multi-model)	RAG I: retrieval basics, embeddings	Build a tiny RAG over 5–10 docs
8	RAG II: evaluation & guardrails	Multimodal I: text + image	Add answer-citing and refusal policies
9	Multimodal II: text + image + audio	Bias & ethics; safety and red-teaming	Draft a safety checklist for your app
10	Data augmentation & RLHF concepts	Model evaluation & feedback loops	Create a synthetic QA set (20 items)
11	Automation & API integration (agents)	Education & business integrations	Call an LLM via REST; log prompts/outputs
12	Practice set & troubleshooting	Final synthesis & showcase	Present a 3-minute demo