Ruth A.
Sample Changed how my team works
We built a shared prompt library after this course, and new colleagues get up to speed much faster.
Prompting AI assistants well: a practical guide for work
A free course on getting reliable, useful results from ChatGPT, Claude, Gemini and similar assistants, and knowing when not to trust them
Created by Elena Petrova
- Updated September 2026
- English
- Beginner
What you will learn
- Explain in plain words how AI assistants generate text and why they make mistakes
- Write clear prompts with task, context, examples, constraints and format
- Refine results through conversation instead of starting from scratch
- Ask for structured output that is easy to reuse
- Check AI output for errors and reduce made up facts
- Use assistants safely without exposing confidential data
Course content
Understanding and instructing AI assistants
Getting reliable output
- Video: Structured output: tables, lists and JSON 10 min
- Video: Reducing made up facts 11 min
- Article: Asking for sources and citations 9 min
- Article: Checking numbers and calculations 9 min
- Video: Reasoning step by step 9 min
Privacy, safety and good judgement
- Video: What not to paste into an assistant 10 min
- Article: Bias, fairness and sensitive decisions 9 min
- Video: Prompt injection and untrusted content 9 min
- Article: Being transparent about AI use 7 min
Workflows for everyday tasks
- Video: Writing and editing 11 min
- Video: Summarising documents and meetings 10 min
- Video: Research and learning 11 min
- Video: Brainstorming and planning 9 min
- Video: Spreadsheets and data 10 min
Building good habits
- Article: A personal prompt library 8 min
- Video: Custom instructions and projects 9 min
- Video: Choosing the right tool for the job 9 min
- Article: Keeping up without chasing hype 6 min
- Quiz: Final check 5 min
Requirements
- Access to any AI assistant, free or paid
Description
AI assistants such as ChatGPT, Claude and Gemini can draft, summarise, explain, translate, brainstorm and analyse. Used well, they save real time. Used carelessly, they produce confident nonsense, leak sensitive data or create more editing work than they save. The difference is mostly in how you ask and how you check.
This free course teaches practical prompting for everyday work. It is not about secret phrases or magic templates. It is about giving the assistant what it needs to do a good job: a clear task, the right context, examples, constraints and a format, and then reviewing the result with the right amount of scepticism.
What you will learn
- How language models work, in plain words, and why that explains their strengths and failures
- A simple structure for clear prompts
- Giving context and examples effectively
- Iterating in a conversation instead of starting over
- Getting structured output such as tables, checklists and JSON
- Spotting and reducing made up facts
- Protecting confidential and personal information
- Practical workflows for writing, research, data and meetings
Vendor neutral
Everything here works with any major assistant. Features differ between products and change often, so we focus on habits that keep working as models improve.
No technical background is needed. If you can write an email, you can write a good prompt.
Who this course is for
- Office workers who want useful results from AI assistants
- Managers and team leads setting good habits for their teams
- Students and freelancers using AI for writing and research
Instructor
Elena Petrova
AI engineer teaching prompting, LLM APIs and retrieval apps that work in practice
- 4.5 instructor rating
- 22 reviews
- 9,111 students
- 3 courses
- Sample
I build applications on top of large language models: assistants, document search and internal tools that save teams real time. I have learned most of what I teach by shipping features, watching them fail in odd ways and fixing them.
My courses are hands on and vendor neutral. We use the APIs from several model providers, compare results, and write small evaluations so you can tell whether a change helped.
I care about the unglamorous parts: cost, latency, privacy and knowing when a plain search or a simple rule is better than a model.
Student reviews
Monica D.
Sample Practical and free
The five part prompt structure changed my results immediately. My emails need far fewer edits now.
Arjun K.
Sample Finally useful meeting summaries
Asking for decisions, owners and deadlines instead of a summary made a huge difference for my team.
Emeka C.
Sample Clear mental model
Understanding that fluency is not truth was the most useful idea in the whole course.
Teresa G.
Sample Great for beginners
Clear explanations without jargon. I would have liked more examples for spreadsheet work.
Wei L.
Sample The privacy section matters
We updated our team guidelines after the lesson on what not to paste. Very sensible advice.
Jonah P.
Sample Balanced and realistic
No hype, just habits that work. The hallucination lesson should be shown to everyone at work.
Fatima H.
Sample Too basic for heavy users
I use AI every day, so much was familiar. Still, the prompt injection lesson was new to me.
Keep exploring
All AI Tools and Prompting courses
More in AI Tools and Prompting
Free
Enroll now