Works with ChatGPT, Claude, Copilot & all other AI tools

Prompt engineering: better AI output with the right technique

Practical in-company training at multiple levels. Learn how to effectively instruct AI models, from first steps to advanced prompt techniques and automation. Applicable to every AI tool.

Half day or full day
Max. 12 participants
Online or on-site
Model-agnostic
★★★★★Rated 4.9 · over 15 years of experience in practical training
Prompt engineering training at Mellaart Trainingen
Why prompt engineering pays off

The quality of your prompt determines the quality of your output

Those who learn how AI models work and how to instruct them effectively get structurally better results, with every tool, every day.

75%
better output with structured prompt techniques
faster to reach the desired AI output
40%
time saved on writing and analysis tasks
9/10
users write suboptimal prompts without training
The programme

5 training programmes, from essentials to specialisation

Choose the level or audience that suits your team, or let us put together a tailored programme.

Essentials  ·  Half day
1

How does an AI model work?

A clear introduction to language models and why the way you ask questions matters so much.

  • What does an AI model actually do under the hood?
  • Why does the same model sometimes give better, sometimes worse answers?
  • Tokens, temperature and context: what you need to know
  • Differences between ChatGPT, Claude, Copilot and Gemini
2

The anatomy of a good prompt

Discover which ingredients make a prompt effective and how to combine them.

  • Role, task, context, format and examples
  • Formulating clearly and specifically
  • Providing context: who, what, for what purpose and for whom
  • Format instructions: lists, tables, steps, length
  • Common beginner mistakes and how to avoid them
3

First prompt techniques

Simple but powerful techniques that deliver immediate results.

  • Zero-shot vs. few-shot: providing examples
  • Iterating: following up, adjusting and refining
  • Role-play: giving the AI a persona
  • Negative instructions: saying what you do not want
  • Setting tone and style
4

Working critically with AI output

Recognising good output and building trust without being naive.

  • What are hallucinations and how do you spot them?
  • Verifying and adjusting output
  • When do you trust AI and when do you not?
  • Privacy and data security: what do and don't you share?
Advanced  ·  Full day
1

Prompt frameworks

Systematic approaches for consistent, high-quality output.

  • RACE method: Role, Action, Context, End result
  • CO-STAR framework: Context, Objective, Style, Tone, Audience, Response
  • RTF method: Role, Task, Format
  • When to use which framework?
  • Adapting frameworks to your own work practice
2

Chain-of-thought & reasoning

Getting AI to tackle multi-step problems and reason step by step.

  • Chain-of-thought prompting: "think step by step"
  • Making intermediate steps visible and verifiable
  • Breaking down complex analyses into subtasks
  • Self-consistency: comparing multiple solutions
  • When is chain-of-thought useful?
3

Few-shot & examples

The power of examples: steering output precisely with concrete references.

  • One-shot, few-shot and many-shot techniques
  • Selecting and structuring good examples
  • Copying style, tone and format via examples
  • Negative examples: showing what you don't want
4

Prompt chaining & workflows

Linking multiple prompts into a streamlined AI workflow.

  • Output of prompt A as input for prompt B
  • Iterative refinement in multiple steps
  • Modularity: reusable prompt building blocks
  • Pitfalls with chained prompts and how to avoid them
5

Building a prompt library

Systematically capturing your best prompts and sharing them with colleagues.

  • What makes a prompt reusable?
  • Building in variables and placeholders
  • Prompt templates for recurring tasks
  • Managing and versioning your prompt library
  • Practical case: build your own team library
Technical  ·  Full day
1

LLM architecture & parameters

Understand how language models work to write better system instructions.

  • Transformer architecture in understandable terms
  • Setting temperature, top-p and max tokens
  • System prompts vs. user prompts
  • Model selection: when do you choose which model?
  • Context window: what fits and what doesn't?
2

API usage & integrations

Calling AI models via the API and integrating them into your own tools.

  • REST API calls: structure and authentication
  • Chat completions vs. completions
  • Streaming and asynchronous processing
  • Error handling and retry logic
  • Managing costs: counting and optimising tokens
3

System instructions & personas

Building robust system prompts for consistent AI assistants.

  • Writing system instructions that stick
  • Setting role boundaries and behaviour rules
  • Providing knowledge bases and fixed context
  • Preventing edge cases and unexpected behaviour
  • Testing and refining instructions
4

RAG basics & context injection

Dynamically adding relevant information to prompts for more accurate output.

  • What is Retrieval-Augmented Generation (RAG)?
  • Chunking and embedding documents
  • Injecting search results into the prompt
  • Monitoring retrieval quality
  • RAG vs. fine-tuning: when to use which?
5

No-code automation

Building AI workflows without programming via Make, Zapier and Power Automate.

  • Embedding prompts in no-code flows
  • Automatically processing emails, forms and data
  • Conditional logic based on AI output
  • Security and governance in automated workflows
Communications & Content
1

Creating content with prompts

AI as a creative writing partner for all forms of content.

  • Writing blog posts, newsletters and articles
  • Tailoring social media copy per platform
  • Generating variants and A/B texts
  • Developing creative concepts with AI
2

Maintaining tone of voice

Preserving your house style and brand identity in AI-generated texts.

  • Including a style guide in the prompt
  • Using example texts as reference
  • Maintaining consistency across multiple authors
  • Adapting AI output without the "AI feel"
3

Rewriting, editing & translation

Improving, simplifying or translating existing texts with AI.

  • Rewriting texts for a different audience
  • Converting jargon into plain language
  • Summarising long documents and reports
  • Multilingual writing and checking for nuance
4

SEO content & metadata

Accelerating search engine optimisation with smart prompt techniques.

  • Supporting keyword research with AI
  • Writing SEO-optimised texts and titles
  • Generating meta texts, alt texts and structured data
  • Planning content with an AI-powered editorial calendar
Management & Analysis
1

Analysis & decision-making

AI as an analytical thinking partner for complex questions and choices.

  • Working out SWOT, PESTLE and scenario analyses
  • Drawing up pro/con analyses and decision matrices
  • Exploring multiple perspectives
  • Recognising and correcting biases in AI output
2

Summarising & synthesising documents

Processing large amounts of information quickly into actionable insights.

  • Summarising reports, policy documents and minutes
  • Distilling patterns and conclusions from multiple sources
  • Writing management summaries and executive summaries
  • Assessing quality and completeness of AI summaries
3

Reporting & presentations

Articulating management information faster and more sharply.

  • Preparing quarterly reports and progress overviews
  • Writing presentation texts and speaker notes
  • Sharpening boardroom communication with AI
  • Formulating KPI explanations and dashboard texts
4

AI strategy & governance

Promoting and monitoring responsible AI use across the organisation.

  • Assessing AI opportunities and risks for your organisation
  • Drafting guidelines and AI policy for employees
  • Guiding employees in responsible AI adoption
  • Building the business case for AI investments
Your benefits

Why choose this prompt engineering training?

No abstract theory, but direct impact on how your team uses AI tools in everyday practice.

Model-agnostic

The techniques you learn work with ChatGPT, Claude, Microsoft Copilot, Gemini and future AI tools. Invest in knowledge that doesn't become outdated with every new tool release.

Immediately applicable

You practise with concrete challenges from your own work environment. Every participant leaves with a personal prompt library full of templates they can use that same day.

Tailored to your level

From employees with no AI experience to developers who want to integrate models into applications: every programme is aligned with the knowledge level and work practice of your team.

Critical & responsible

As well as effectiveness, you cover the limits of AI: recognising hallucinations, verifying output, protecting privacy and responsibly embedding AI in work processes.

Experienced AI trainer

Over 15 years of training experience, combined with in-depth and up-to-date knowledge of the AI landscape. Practical, clear and jargon-free, for every audience.

Materials & follow-up

After the training, every participant receives a summary of all techniques covered, a tailored prompt library and the option to follow up with questions afterwards.

For whom

Is the prompt engineering training right for you?

The training is suitable for any professional who uses (or wants to start using) AI tools in their work.

Employees & knowledge workers

Learn to instruct ChatGPT, Copilot or Claude more effectively for writing, summarising, analysing and structuring. Save time and improve the quality of your AI output immediately.

Managers & team leads

Use AI as an analytical thinking partner for better decision-making, faster reporting and more effective communication. Learn how to guide your team in responsible AI use.

Communications & marketing

Generate better content faster without losing your own voice. Learn how to provide AI with style guides and tone of voice so every text reflects your brand.

IT professionals & developers

Learn to write system instructions and API calls, embed prompts in no-code flows and apply RAG fundamentals. For everyone who wants to structurally integrate AI.

Trainers & educators

Integrate AI as a learning and development tool. Learn how to use AI for creating training materials, quiz questions, practice cases and personalised learning pathways.

Anyone who wants more from AI

No technical background required for the Essentials training. Those who already work regularly with AI and want structurally better results can choose Advanced or an industry-specific variant.

Practical information

Everything you need to know

Approach

The training combines short theory segments with extensive hands-on exercises. Participants work directly with AI tools on scenarios from their own work practice. This way they leave with concrete knowledge, proven techniques and a personal prompt library.

Choosing or combining levels

Not sure which level suits your team? In a no-obligation introductory conversation we map out the knowledge level and preferences. We can also combine topics from different levels into a tailored programme that fits your organisation precisely.

Prior knowledge per level

Essentials: No prior knowledge required. Basic computer experience and some familiarity with AI tools is sufficient.
Advanced: You already have some experience with AI tools and want to work more systematically and effectively.
Technical: You have programming experience (Python or JavaScript) and want to integrate AI models into applications and workflows.

Flexible duration

The Essentials training is a half day. The Advanced and Technical programmes are set up as a full day. Industry-specific variants (Communications & Content, Management & Analysis) can be combined with a level training. We always align the duration with your preferences.

Materials & follow-up

After the training, every participant receives a comprehensive summary of all techniques covered, a tailored prompt library and an overview of recommended follow-up resources. For questions after the training, you can always get in touch.

Combination with other AI trainings

The prompt engineering training connects seamlessly with our ChatGPT training, Claude training and Microsoft 365 Copilot training. Would you like a broad AI programme for your organisation? We are happy to put together a tailored programme.

Our approach

From request to result

Every prompt engineering training is carefully prepared and aligned with your organisation.

1

Introduction

A short conversation to understand your requirements, audience, knowledge level and current AI use. No obligation.

2

Tailored programme

We choose the right level and put together a programme with examples from your own work practice.

3

Training delivery

Interactive, practical and engaging. Online or on-site. Directly applicable techniques and prompt templates.

4

Follow-up

Summary, prompt library and the option to follow up with questions. So what you've learned sticks and grows.

Experiences

What participants say

★★★★★

"I had been using ChatGPT for months, but after this training I realised how much I was missing. The prompt frameworks are a revelation: my output has immediately become so much better."

Participant
Prompt Engineering Advanced · Government
★★★★★

"Finally a training that explains why certain prompts work and others don't. Very practical, with exercises that directly connect to our work. As a team we have really taken a step forward."

Participant
Prompt Engineering Essentials · Communications
★★★★★

"The technical deep dive was exactly what I was looking for. System instructions, RAG fundamentals and API integration, all explained clearly and accessibly with practical examples."

Participant
Prompt Engineering Technical · IT & Software
Frequently asked questions

Everything about the prompt engineering training

Prompt engineering is the art and science of formulating the right instructions to an AI model to get the best possible output. With the right techniques (such as providing context, assigning a role, giving examples or reasoning step by step) you get significantly better results from ChatGPT, Claude, Copilot or other AI tools.

The techniques you learn are applicable to all major language models: ChatGPT, Claude, Microsoft 365 Copilot, Gemini and others. Prompt engineering is model-agnostic: the principles work everywhere. During the training we preferably use the AI tools your team already uses in practice.

The ChatGPT training and Claude training focus on the specific tool: all features, interfaces and capabilities of that one tool. The prompt engineering training focuses on the underlying techniques: techniques you can then apply to any AI tool. The trainings complement each other well.

The Essentials training is a half day. The Advanced and Technical programmes are set up as a full day. Industry-specific variants can be combined with a level training. We always align the duration with your preferences and the knowledge level of the group.

That is not a problem. In the introductory conversation we discuss the knowledge level of the group and put together a programme that is challenging and relevant for everyone. Our trainers are experienced in flexibly handling level differences within a group.

Yes, all our prompt engineering trainings are available in-company, at your location or online. Content, examples and exercises are fully tailored to your organisation and industry.

Ready to get started?

Book the prompt engineering training for your team

Fill in the form on our contact page or call us directly. We will get back to you within two working days to discuss your requirements. No obligation.