Built for complete beginners who want to become more confident working with data. You will complete three practical projects across: - Retail - Health - Finance In each one, you will use ChatGPT, Claude or Gemini alongside Excel or Google Sheets to understand an unfamiliar dataset, spot what the AI missed and verify one important figure by hand. Each project takes about 45–60 minutes. No coding, no installation and no previous experience required. You can start with any industry: - Retail — What Is Really in the Sales File? - Health — Reading the A&E Monthly Return - Finance — Understanding Customer Complaints When you finish, share three short lines in the community: The one thing I found: The number I verified: The thing that looked wrong but wasn’t: Start here: Classroom → Real Data, Real Decisions — Level 1

What you will learn In this playbook, you will learn how to design an AI workflow that uses different models for different tasks instead of relying on one model for everything. You will decide: Which tasks need a fast model Which tasks need stronger reasoning Which tasks need verification Where human approval is required How to control cost, quality and risk How information should move between stages You will finish with a complete multi-model agent workflow plan for one real task. Who this playbook is for This playbook is suitable for learners who: Use more than one AI platform Want to build simple AI-agent workflows Need to balance quality and cost Want to reduce errors in AI-generated work Are interested in routing tasks between different models Want to add verification before trusting AI outputs You do not need coding experience. The main activity can be completed using ordinary AI tools and a document, spreadsheet or diagram. Estimated time: Approximately 45 to 60 minutes. Level: Beginner to Intermediate The workflow-design activity is beginner-friendly. Connecting models through APIs, automation platforms or agent frameworks is an intermediate extension. 1. What is a multi-model AI workflow? A multi-model workflow uses different AI models for different parts of a task. Instead of asking one model to: Understand the request Research the topic Produce the answer Check the facts Review the quality Approve the final result the work is divided into stages. Each stage is assigned to the model or tool best suited to that job. A simple example might be: A fast model classifies the request. A stronger model creates the main answer. A second model checks accuracy and completeness. A human reviews the final result. This is a basic AI-agent workflow. 2. Why not use one model for everything? Using one model is often the easiest approach. It may be suitable when: The task is simple. The output is low risk. Speed matters more than perfection. The work is completed only occasionally.

What you will learn In this playbook, you will learn how to assess whether your computer is suitable for fine-tuning an open-weight language model. You will identify: Your available hardware Your GPU memory A suitable model size Whether LoRA or QLoRA is appropriate A realistic sequence length and batch size Whether you should train locally, reduce the configuration or use a cloud GPU You will finish with a short fine-tuning feasibility report for one model. Who this playbook is for This playbook is suitable for learners who: Are curious about fine-tuning language models Want to experiment with local AI Have heard about LoRA or QLoRA but are unsure where to start Want to avoid downloading a model that their computer cannot train Need a simple way to compare model requirements with available hardware You do not need to be an experienced machine-learning engineer. Some optional steps use Python and the command line. Estimated time: Approximately 45 to 60 minutes. Level: Beginner to Intermediate The hardware-planning activity is beginner-friendly. Running the optional local estimator and generating a training recipe introduces intermediate technical skills. 1. What does fine-tuning mean? A pre-trained language model has already learned patterns from a large collection of text. Fine-tuning continues the model’s training using a smaller, more focused dataset. This can help the model become better at: Following a specialised format Using a particular tone Answering questions within a narrow domain Producing organisation-specific outputs Completing a repeated task more consistently Adapting to examples from a specialist workflow Examples include: A customer-support assistant trained on approved response examples A coding model adapted to a particular programming framework A learning assistant trained to follow an assessment rubric A local model adapted to a company’s report-writing style An assistant trained to convert notes into a fixed document structure

What you will learn In this playbook, you will learn how to test whether an AI Skill genuinely improves an AI model’s performance. You will compare the same task in two conditions: Test A: The AI completes the task without the Skill. Test B: The AI completes the task with the Skill. You will then evaluate both results using clear criteria rather than relying on which answer simply “looks better”. Who this playbook is for This playbook is suitable for beginners who: Use ChatGPT, Claude, Gemini or another AI assistant. Create reusable prompts, Custom GPT instructions, Gems or Agent Skills. Want more consistent AI results. Need a simple way to test whether their instructions are working. Want to improve an AI workflow without guessing. You do not need coding experience. Estimated time: Approximately 35 to 50 minutes. Level: Beginner Tools you will need: Choose one AI assistant: ChatGPT Claude Gemini Microsoft Copilot Another text-based AI assistant You will also need somewhere to record your results: A Word or Google document A spreadsheet A notes application The evaluation table included in this playbook For the beginner activity, you do not need to install any software. An optional advanced activity later in the playbook introduces the open-source agent-skills-eval project. 1. What is an AI Skill? An AI Skill is a reusable set of instructions, examples, rules and resources that helps an AI perform a specific task. A Skill might teach an AI how to: Review a CV. Analyse a spreadsheet. Write a lesson plan. Assess learner work. Research a topic. Create a presentation. Review another AI agent’s code. Produce a report using a required format. A Skill may be stored in different ways. For example: Custom instructions A Custom GPT A Gemini Gem A saved master prompt A SKILL.md file A reusable workflow inside an AI application The important idea is that the Skill should provide more than a one-time prompt. It should give the AI a consistent method to follow.

You will learn how to use AI to transform unstructured notes from a meeting, lesson, conversation or planning session into clear actions. This playbook is designed for beginners who: - Take notes but rarely return to them. - Forget actions agreed during meetings. - Struggle to separate information from decisions. - Need a simple way to identify deadlines and follow-ups. - Want to reduce the time spent rewriting notes. Estimated time: 25 to 35 minutes. Tools Use any general AI assistant, such as: - ChatGPT - Gemini - Claude - Microsoft Copilot You will also need a set of notes. These may come from: - A meeting. - A lesson. - A webinar. - A planning discussion. - A personal voice-note transcript. - A brainstorming session. - A project update. Remove private, confidential or identifying information before pasting notes into an AI tool. The problem Notes often mix several types of information: - Background information. - Ideas. - Decisions. - Questions. - Tasks. - Deadlines. - Names. - Risks. - Follow-up points. Consider these notes: Website launch discussed. Ahmed will check the images. Need to confirm the price page before Friday. The contact form may still be sending messages to the old email address. Consider adding testimonials later. Sarah asked whether the training page should include a downloadable guide. Review again next Tuesday. A person reading these notes must work out: - What was decided? - What needs doing? - Who owns each action? - Which deadline matters? - What remains uncertain? - What can wait? AI can help structure this information, but you must check the result. Step 1: Choose a real set of notes Select notes that contain at least: - One action. - One decision or important point. - One question or uncertainty. - One deadline or follow-up date. The notes do not need to be neat. Messy notes are suitable because organising them is part of the skill. Step 2: Remove sensitive information