Best broad prompting base: Learn Prompting. Sander Schulhoff's guide to prompting patterns and risks. Start here to understand the reusable mechanics behind writing prompts.
Best current OpenAI prompting reference: GPT-5.5 prompting guide. Official OpenAI guide for GPT-5.5 prompting adjustments. Use it when writing workflows depend on current OpenAI model behavior.
Best Claude prompting reference: Anthropic Prompt Engineering Overview. Official Anthropic prompting guidance. Use it when your drafting, editing, or research workflows run through Claude.
Writing with AI needs editorial control
AI can help with outlining, drafting, editing, summarising, research, repurposing, and style exploration. The writer still owns taste, structure, accuracy, voice, and final judgement.
Learn Prompting is a strong base for prompt mechanics. Official OpenAI and Anthropic prompting docs matter when the writing workflow depends on a specific model's behavior.
Build a repeatable writing workflow
A useful AI writing workflow separates research, notes, outline, draft, critique, rewrite, fact check, and final edit. Asking a model for a complete finished piece in one step usually creates bland or unreliable work.
Look for resources that teach revision and verification. The best AI writing practice makes the writer sharper rather than outsourcing judgement to the model.
Recommended courses and resources
-
ChatGPT Prompt Engineering for Developers
Short course · DeepLearning.AI · Beginner
You want a short, structured intro to prompting for software tasks.
-
Learn Prompting
Guide · Sander Schulhoff · Beginner to intermediate
You need a broad prompt engineering reference.
-
Prompt Engineering Guide
Guide · DAIR.AI · Beginner to advanced
You want examples of prompting techniques and patterns.
-
GPT-5.5 prompting guide
Guide · OpenAI · Intermediate
You want the official prompting adjustments for GPT-5.5 instead of reusing older GPT-4 or GPT-5 prompt habits unchanged.
-
OpenAI model optimization
Guide · OpenAI · Intermediate
You need a practical optimization loop across prompt changes, evals, and fine-tuning rather than guessing which knob to turn next.