A curated and structured learning path for prompt engineering.
This repository is part of the Awesome Learn ecosystem. It is designed to help learners move from basic prompting to advanced prompt systems, workflows, and evaluation.
Prompt engineering is the practice of designing inputs for large language models to produce more accurate, useful, and reliable outputs.
It sits at the intersection of language, logic, systems thinking, and human-computer interaction.
Prompt engineering matters because better prompts improve:
- output quality
- consistency
- reasoning depth
- task efficiency
- workflow reliability
It is one of the most practical AI skills today.
Prompt engineering spans several connected areas:
- Prompt structure
- Context management
- Role prompting
- Chain-of-thought prompting
- Few-shot prompting
- Retrieval-augmented prompting
- Agent workflows
- Evaluation and benchmarking
- Safety and guardrails
Start here:
- What large language models are
- Tokens and context windows
- Temperature and sampling
- System prompts vs user prompts
- Why specificity matters
Core concepts:
- instruction clarity
- constraint framing
- output formatting
- examples as guidance
Build stronger prompts:
- Zero-shot prompting
- Few-shot prompting
- Persona prompting
- Step-by-step prompting
- Decomposition prompts
- Iterative refinement
Focus on:
- reducing ambiguity
- improving consistency
- debugging prompts
Move into more complex systems:
- Chain-of-thought
- Tree-of-thought
- Self-critique loops
- Reflection prompting
- Tool-use prompting
- Multi-agent prompting
- RAG prompt design
Learn when each technique helps.
Apply prompt engineering to:
- Writing
- Coding
- Research
- Summarization
- Knowledge management
- Data extraction
- Decision support
- Workflow automation
Practicality is where skill compounds.
Suggested projects:
- Build a reusable writing prompt library
- Create an AI research assistant workflow
- Design a customer support prompt system
- Build structured extraction prompts
- Create evaluation prompts for output scoring
- Build a prompt testing harness
- OpenAI Prompting Guide — foundational prompt principles
- Anthropic Prompt Engineering Docs — practical Claude prompting patterns
- Google Prompt Design Guide — Gemini-oriented workflows
- Learn Prompting (learnprompting.org) — structured free curriculum
- DeepLearning.AI short courses — practical AI workflow courses
- OpenAI Playground
- Anthropic Console
- LangChain
- Flowise
- Promptfoo
- Future AGI
Key areas to explore:
- in-context learning
- reasoning benchmarks
- hallucination reduction
- evaluation frameworks
- agent reliability
After prompt engineering, consider:
- AI workflow design
- Retrieval systems
- Agent architecture
- Evaluation systems
- AI governance
- Human-AI collaboration
Prompt engineering is often the entry point, not the endpoint.
Please read CONTRIBUTING.md before submitting changes.
This list is licensed under Creative Commons Zero v1.0 Universal.