AI Foundations Bootcamp

Mental Models of LLMs

Tokens, context, sampling, and why models hallucinate; how to design around it.

Learning outcomes

- Explain tokens/context windows and their implications. - Use temperature/top_p intentionally for d…

Project

Write a one-page 'LLM Operating Manual' for your own use, including do/don't prompt patterns, eval c…

Prompting Patterns That Work

System vs user instructions, constraints, examples, rubrics, and self-check prompts.

Learning outcomes

- Apply 8 repeatable prompt patterns to real problems. - Design output schemas and enforce them. - B…

Project

Build a prompt library: 10 templates for summarization, extraction, planning, critique, and role-bas…

AI Workflow Design

When to use single-shot vs multi-step workflows, tool use, and guardrails.

Learning outcomes

- Break tasks into reliable steps. - Choose tools and inputs/outputs for each step. - Add guardrails…

Project

Design a 5-step workflow for 'content to landing page' and document inputs/outputs, validation rules…

Evaluation & Quality

Rubrics, gold sets, regression tests, and measuring improvements.

Learning outcomes

- Write clear scoring rubrics. - Build a small evaluation set and track quality. - Run A/B compariso…

Project

Create a 30-item eval set for one task you care about; implement a rubric and record baseline scores…

Safety, Ethics, and Policy Basics

Privacy, data handling, bias, and safe deployment practices.

Learning outcomes

- Identify data you should never send to third parties. - Create a lightweight safety policy for you…

Project

Draft an 'AI Use Policy' for a small business: allowed use cases, prohibited data, review requiremen…