terminal

Prompt Engineering Mastery

The art & science of getting the best out of LLMs — from first prompt to production
Co-Created by Kiran Shirol and Claude
TopicsFoundationsReasoningControlReal-WorldProduction
home Learning Portal play_arrow Start Learningsummarize Key Insightsdictionary Glossary14 chapters · 5 sections
Section 1

Foundations — Stop Prompting Like a Google Search

How prompts work, the anatomy of a great prompt, and few-shot learning.
1
lightbulb
How Prompts Actually Work
Tokens, probabilities, and why vague prompts get vague answers.
arrow_forward Learn
2
build
Anatomy of a Great Prompt
Role, Context, Task, Format, Constraints — the 5 building blocks.
arrow_forward Learn
3
format_list_numbered
Zero-Shot vs Few-Shot
When to give examples, how many, and how to pick good ones.
arrow_forward Learn
Section 2

Reasoning — Make the Model Think

Chain-of-thought and advanced reasoning patterns.
4
psychology
Chain-of-Thought Prompting
“Think step by step” — the difference between wrong and right on hard problems.
arrow_forward Learn
5
account_tree
Advanced Reasoning Patterns
Tree-of-thought, self-reflection, decomposition, and verification chains.
arrow_forward Learn
Section 3

Control — Consistent, Reliable Output

System prompts, structured output, and reusable patterns.
6
person
System Prompts & Personas
Personality, tone, and behavioral boundaries for specialized assistants.
arrow_forward Learn
7
data_object
Output Formatting & Structured Data
JSON, XML, markdown, tables — reliable structured output every time.
arrow_forward Learn
8
pattern
Prompt Patterns & Reusable Templates
Critic, Decomposer, Verifier, Persona Chain — battle-tested patterns.
arrow_forward Learn
Section 4

Real-World Complexity

Code, RAG, and multi-turn conversation design.
9
code
Prompting for Code
Generation, debugging, refactoring — why “write a function” fails.
arrow_forward Learn
10
description
RAG & Context Injection
Feeding external documents into prompts effectively.
arrow_forward Learn
11
forum
Multi-Turn & Conversation Design
Context across turns, steering, and progressive refinement.
arrow_forward Learn
Section 5

Mastery — Production-Ready

Tool use, evaluation, debugging, and the prompt engineer’s toolkit.
12
build
Tool Use & Function Calling
Tool descriptions ARE prompts — the same principles apply.
arrow_forward Learn
13
bug_report
Evaluation, Debugging & Pitfalls
Systematic debugging, LLM-as-judge — prompts are software, test them.
arrow_forward Learn
14
handyman
The Prompt Engineer’s Toolkit
DSPy, prompt chaining, auto-optimization, and your decision tree.
arrow_forward Learn
explore

Explore Related Courses

token
How LLMs Work
Transformers & Attention
smart_toy
Agentic AI
Planning, Memory & Tool Use
search
RAG
Retrieval-Augmented Generation