Prompt Engineering Fundamentals

One instructor-led day, platform-neutral and built around everyday workplace tasks.

Explains what is happening behind the screen — tokens, context windows, how a model generates a response — so prompting decisions are informed rather than superstitious.

• Moves from single prompts to structured prompts to a custom assistant you build yourself.
• Treats verification, privacy and responsible use as part of the work, not an appendix.

• Professionals who want to use generative AI more effectively in their daily work.
• Employees, managers, educators, content creators, analysts, marketers and project teams.
• Anyone improving productivity, communication, research or problem-solving with AI.
• Teams asked to adopt AI who need to judge its output rather than accept it.

• No previous experience with artificial intelligence, programming or prompt engineering is required.
• Basic computer and internet skills.
• Experience using common workplace applications.
• Access to an AI chatbot such as ChatGPT, Google Gemini or Microsoft Copilot.

By the end of this course, participants will be able to:

• explain how generative AI and large language models produce responses
• compare common AI chatbots and select an appropriate tool for a defined task
• construct effective prompts using clear objectives, relevant context, constraints, examples and output requirements
• apply structured prompting techniques to improve the relevance, consistency and usability of AI-generated results
• evaluate AI output for accuracy, bias, misinformation and potential risk
• apply privacy, security and responsible AI practices when using AI in the workplace
• create, test and safely share a custom AI assistant for a practical business use case

• Module 1. Understanding your new digital assistant
  • Generative AI can help with writing, research, analysis, brainstorming and many everyday tasks. To use it well it helps to understand what happens behind the screen. Large language models break prompts and responses into tokens, and token count affects cost, speed and response length. Each model also has a context window that limits how much it can consider at once. The module covers how models learn and generate, how they handle text and images, where they fall short, and why human judgment still matters.
• Module 2. Designing effective prompts
  • Results depend heavily on the instructions and information the model receives. A strong prompt states the goal, supplies useful context and describes what a good response looks like. Examples can be typed in or uploaded as documents and images, and several examples can demonstrate a pattern through few-shot prompting. AI can help draft and improve a prompt; testing and adjusting is what makes the result usable.
• Module 3. Choosing your AI assistant
  • Assistants differ in what they can do. Some work across text, images, audio and files; others connect to workplace applications or specialize in particular tasks. Models, features, usage limits, privacy controls and pricing all vary. Comparing them makes it easier to pick an assistant that fits the task, the information involved and the organization's requirements.
• Module 4. Structuring prompts for reliable results
  • When a request carries several instructions, structure tells the model what belongs together and what matters most. Organize prompts with headings, lists, delimiters, Markdown and XML-style tags; set constraints, define an output format, give examples and break complex tasks into steps. The result is less confusion, more consistency and less time spent revising.
• Module 5. Think before you trust AI
  • AI can answer confidently and still be incomplete, outdated, biased or wrong. Learn why that happens and how to catch it before you use or share the content: checking claims, verifying sources, looking for bias and protecting sensitive information — and knowing when to rely on your own expertise, ask someone else, or follow your organization's AI policy.
• Module 6. Protecting people and data
  • Using AI responsibly means weighing its effect on people, information and decisions. Work through fairness, privacy, security, transparency and accountability: how to handle different kinds of data, who is responsible for decisions involving AI, and how to judge ethical and legal risk — then choose safeguards that hold from tool selection through ongoing review.
• Module 7. Building your own AI assistant
  • Build an assistant for a specific audience, task or workflow without writing code. Compare custom GPTs, Gemini Gems and NotebookLM, pick a practical use case and build a simple assistant: clear instructions, trusted information, sensible boundaries, and a test of how it responds. Finish with how to maintain and share it safely inside your organization.

Nine guided activities run through the day:

• Explore what AI chat applications can do
• Building better AI prompts
• Creating images with AI
• Structuring prompts
• Context, ambiguity and AI interpretation
• Prompt guardrails
• Analyze customer feedback with generative AI
• Create a custom GPT (ChatGPT)
• Create a Gemini Gem (Google Gemini)

• Practical AI for Business
• AI at Work: Safe, Secure & Compliant
• Introduction to Agentic AI