Prompt engineering at the intermediate level is less about memorizing trick phrases and more about learning a repeatable workflow: define the task clearly, add the right context, constrain the output format, provide examples, test variations, and evaluate results systematically. For someone who already uses ChatGPT, Google Gemini, and Claude day to day, the best free courses are the ones that move beyond basic prompting and teach prompt patterns, iteration, context management, structured outputs, evaluation, and real-world workflows. These courses were selected because they are current, from reputable providers, and either fully free or available with a genuine free audit/free access path. ([coursera.org](https://www.coursera.org/learn/prompt-engineering?utm_source=openai)) The strongest overall recommendations are Vanderbilt/Coursera’s Prompt Engineering for ChatGPT for practical prompt patterns, DeepLearning.AI’s ChatGPT Prompt Engineering for Developers for concise hands-on practice, Microsoft Learn’s prompt engineering modules for structured fundamentals, Google’s Gemini prompt-design resources for model-specific best practices, and Anthropic’s Claude learning resources for long-context and reliability strategies. Together, they map directly onto the tools you already use: ChatGPT/OpenAI for prompt patterns and application design, Gemini for concise and well-structured prompting, and Claude for long-context reasoning and robust task instructions. ([coursera.org](https://www.coursera.org/learn/prompt-engineering?utm_source=openai))
Prerequisites: You should already be comfortable using at least one major AI assistant for everyday tasks, writing multi-step instructions, revising prompts based on outputs, and understanding basic concepts like context windows, hallucinations, and few-shot examples. No heavy programming background is required for most of these, though a little API or notebook familiarity helps for the DeepLearning.AI and Google lab resources.
Estimated time: Approximately 30-40 hours total
Start with Microsoft Learn’s prompt engineering fundamentals to establish a clean vocabulary for prompt components and iteration. Next, take DeepLearning.AI’s ChatGPT Prompt Engineering for Developers for fast hands-on practice, then Vanderbilt’s Prompt Engineering for ChatGPT to build a broader library of prompt patterns and reusable workflows. After that, use AI Prompting for Everyone to generalize those habits across the exact tools you already use—ChatGPT, Claude, and Gemini. Then branch into provider-specific depth: take Google’s Vertex AI Prompt Design to sharpen Gemini prompting, and Anthropic Academy to learn Claude-specific strategies like long-context prompting and more reliable task design. Finish with Prompt Engineering for LLMs on Coursera to consolidate everything into a more systematic, production-minded view of prompt design, evaluation, memory, and scalable workflows.
One of the best-known practical prompt engineering courses, this teaches reusable prompt patterns, few-shot prompting, structured prompting, and prompt-based application design. It is ideal for learners who already use ChatGPT and want a more systematic way to get better outputs from ChatGPT and similar LLMs.
Topics: prompt patterns, few-shot prompting, structured outputs, LLM workflows, ChatGPT
Go to course →A compact but high-value course taught by Andrew Ng and Isa Fulford that covers prompt engineering best practices for summarizing, inferring, transforming, expanding, and chatbot building. It is especially useful if you want hands-on practice and a clearer mental model for how to iterate prompts in ChatGPT and OpenAI-style workflows.
Topics: iterative prompting, application prompting, summarization, classification, chatbot design, ChatGPT
Go to course →This course is explicitly aimed at people who use tools like ChatGPT, Claude, or Gemini in daily work and want significantly better results. It is a strong bridge from casual prompting to more deliberate prompting across multiple AI assistants, making it especially relevant for your current tool stack.
Topics: cross-tool prompting, workflow prompting, prompt iteration, ChatGPT, Claude, Gemini
Go to course →This Microsoft Learn lesson explains prompt components, iteration, and best practices in a clean, structured format. It is a good reinforcement resource for intermediate learners who want a concise framework they can apply across ChatGPT, Claude, Gemini, and Copilot-style assistants.
Topics: prompt structure, best practices, iteration, generative AI fundamentals
Go to course →This newer Coursera course goes beyond basic prompting into evaluation, optimization, long-context management, conversational memory, multimodal prompts, and scalable prompt systems. It is a strong intermediate next step for learners who want more rigor than one-off prompt tips and are ready to think in terms of reusable systems.
Topics: prompt evaluation, context management, multimodal prompts, memory, LLM systems
Go to course →This Google lab teaches prompt engineering with Gemini-focused best practices such as conciseness, specificity, task definition, and use of examples. It is especially valuable if you use Gemini and want guidance from Google on how to structure prompts for reliable responses.
Topics: Gemini prompting, prompt design, few-shot examples, Google Gen AI SDK, task definition
Go to course →Anthropic’s learning hub includes an interactive prompt engineering tutorial, real-world scenarios, and documentation on maximizing Claude’s performance. It is the best free official resource for learning how prompt engineering changes when working with Claude, especially for long-context tasks, reliability, and evaluation.
Topics: Claude prompting, long-context prompting, evaluation, prompt engineering documentation, real-world scenarios
Go to course →Enter your email and we'll generate your personalized free-course roadmap as a PDF you can keep.
Describe any AI topic and our engine will scan the web to build a personalized free course roadmap you can download.
Build my roadmap →