From Zero to AI Agents
A 4-class journey from no programming background to building AI agents
UC Berkeley Haas AI Course
From Zero to AI Agents
A practical course for learning enough software, AI, LLMs, and agent architecture to reason clearly about modern AI products and build useful prototypes.
5
class decks from orientation through multi-agent systems
2h
roughly per class, with examples students can revisit
0
programming background required before starting
What This Course Is For
AI fluency is becoming a baseline business skill. This course is built for people who want to adapt without pretending they need to become machine learning engineers.
Make AI less intimidating
Build clear mental models before jargon, so modern AI systems feel understandable instead of magical.
Give business students technical leverage
Learn enough software and AI foundation to ask better questions, evaluate tools, and build useful prototypes.
Prepare for AI-native work
Understand workflows, tools, agents, and what should or should not be automated.
Classes
Each class links to the live slide deck. Classes 2-4 also include the session recording.
Class 0: Why This Course Exists
Meet your instructor, understand the AI landscape, and learn what makes this course different.
Class 1: Introduction to Software
What is code? Frontend, backend, databases, cloud platforms, and your first deploy.
Class 2: Introduction to AI
What is AI, how machines learn, and the path from chess to ChatGPT.
Class 3: LLMs in Practice
Prompting theory and practice, reasoning models, structured outputs, RAG, tool use, MCP, skills, and intro agent loops.
Class 4: Building AI Agents
Multi-agent systems, frameworks, Claude Code, and real-world agent architectures.
What Students Said
Anonymous notes from students who came in with different levels of technical background. The goal was confidence, not jargon.
“This was the first AI resource that made me excited to get out of the classroom and continue building on my own machine.”
“You turned a topic that intimidates a lot of people into something the whole community could actually engage with.”
“I feel more confident going into conversations about AI.”
“The materials filled gaps I had from trying to learn AI here and there.”
“I didn’t know where to start, and this class opened that door for me.”
“It helped me understand AI concepts in a way that felt approachable, useful, and not intimidating.”
Prerequisites
None. Bring curiosity and a laptop with a modern browser.
How to Use
Each class is a reveal.js slide deck. Use arrow keys or space to move through slides, press S for speaker notes, O for overview, and F for fullscreen.
Who’s teaching this?
¡Hola! I’m Pepe.
I build products, teach AI, and vibe ship OSS - trying not to let MBA energy or AI sycophancy leak in.