AI Kick Start × AI Kick Start
Agentic Workflows for Knowledge Workers
A 38-lesson AI Kick Start course that starts with plain-language AI basics, builds into practical AI 101, then teaches Codex, Claude, safe agent workflows, reusable skills, automations, and guided product walkthroughs.
0 of 38 lessons complete
16 AI Basics, AI 101, workflow lessons, and walkthroughs included
AI Basics: Start Here
Simple AI foundations first, then business AI 101.
The first lessons use very simple words, short practice tasks, and clear safety rules. Once learners know the basic terms, the existing Careersy AI 101 lessons continue into roadmaps, tools, automation, document AI, repeatable systems, and search visibility.
1
AI Basics: Start Here
AI Basics: What AI Is
AI is computer help that looks for patterns and gives a useful answer. It can help you write, plan, explain, sort, and summarise, but it can still make mistakes.
- Visualisation
- Simple map: question in, AI answer out, human checks before use.
- Learner output
- A one-sentence explanation of AI and a short list of three safe ways to use it.
AI Basics: Start Here
AI Basics: ChatGPT, Claude, and Codex
ChatGPT and Claude are AI chat tools. Codex is an AI tool that can help with project files and code. They overlap, but beginners should choose based on the job.
- Visualisation
- Tool chooser: ask, read, plan, or work on a project.
- Learner output
- A simple tool-choice note for three common jobs.
AI Basics: Start Here
AI Basics: Computer Words You Need
Browser, website, app, download, install, account, sign in, file, and folder are the basic words learners need before setup screens make sense.
- Visualisation
- Plain-language glossary board for setup words.
- Learner output
- A personal glossary of basic computer words.
AI Basics: Start Here
AI Basics: How To Stay Safe
The safest beginner rule is simple: do not share secrets, passwords, payment details, private emails, or sensitive documents unless a trusted person has approved the task.
- Visualisation
- Safe to share vs stop and check.
- Learner output
- A personal AI safety checklist.
AI Basics: Start Here
AI Basics: Your First Simple Prompt
A prompt is the instruction you type. A good beginner prompt names the topic, the audience, and the kind of answer you want.
- Visualisation
- Prompt recipe: topic, audience, output.
- Learner output
- One simple prompt and one improved version.
AI Basics: Start Here
AI Basics: Talking To AI vs Asking AI To Do A Task
Talking to AI gives you an answer. Asking AI to do a task may involve reading, creating, changing, or sending something. That is when you slow down.
- Visualisation
- Answer only vs task with a check point.
- Learner output
- A list of examples labelled answer-only or task.
AI Basics: Start Here
AI Basics: What Permissions Mean
Permission buttons are choices that let a tool see or do something. Beginners should pause, read, and ask what the button allows.
- Visualisation
- Permission words translated into plain English.
- Learner output
- A permission pause checklist.
AI Basics: Start Here
AI Basics: First Safe Mini Project
The first project is a simple checklist. Learners give AI a safe task, read the answer, improve it, and decide what they would use.
- Visualisation
- Mini project path: choose, prompt, read, improve, save.
- Learner output
- A checked weekly checklist made from a simple AI prompt.
Renamed from the AI roadmap guide
AI 101: Build Your First AI Roadmap
Learners pick one real, repeated workflow, score four candidates for impact, effort, data readiness, risk and owner confidence, and leave with a one-page first-win roadmap that names an owner, allowed inputs, a success metric, and a 30-day review checkpoint.
- Visualisation
- Roadmap heatmap showing impact, effort, data readiness, risk, and owner confidence.
- Learner output
- A one-page AI roadmap: first-win workflow, named owner, allowed inputs and data boundary, success metric, key risks, 30-day review checkpoint, and an explicit stop-or-expand decision rule.
Renamed from the best AI tools guide
AI 101: Choose a Starter Tool Stack
Learners map their real weekly work to a short list of jobs, score candidate tools on fit, risk, integration and total cost, then commit to one primary tool, one controlled backup, and a written approval boundary per job - the stack you can actually run, not the longest wishlist.
- Visualisation
- Anchor-platform matrix: ChatGPT, Claude, Microsoft 365 Copilot, and Gemini matched to jobs, ecosystems, boundaries, and budget rules.
- Learner output
- A starter stack scorecard: for each named job, your primary tool, controlled backup, the data it may touch, the monthly cost, and the approval boundary a non-user manager could sign off.
Renamed from the AI automation savings guide
AI 101: Save Hours With Weekly Automation
Learners design a draft-and-review automation for one well-understood weekly task instead of trying to automate a whole role. The lesson teaches the trigger-inputs-actions-draft-review-handover anatomy, how to estimate saved time honestly (including review time), and the staged-trust rule: prove the manual workflow, then the draft path, then widen permission - never the other way round.
- Visualisation
- Before-and-after time bar chart for admin, reporting, research, content, and follow-up work.
- Learner output
- A one-page weekly automation brief: the named task, its trigger, the exact allowed inputs, the draft output and where it is saved, the review rule (who checks what), the failure message, and a conservative saved-time estimate that already subtracts review time.
Renamed from the AI agents explained guide
AI 101: Understand AI Agents
Learners learn the industry definition that separates an agent from a chatbot - autonomous decision-making within a scoped task - then classify their own work as chat, assistant workflow, or agent workflow, and write a permissioned, reviewable agent brief with explicit stop conditions.
- Visualisation
- Agent loop diagram: goal to context to tools to permissions to action to evidence to review and stop.
- Learner output
- A task classification sheet (chat / assistant workflow / agent workflow) plus a safe agent brief for one professional workflow, including its permission scope, evidence requirements, and stop conditions.
Renamed from the choosing AI tools guide
AI 101: Buy Tools Without Wasting Money
Learners run a practical buying filter - problem fit, data boundary, integration, adoption cost, exit path, and a time-boxed pilot decision - before approving any AI subscription, so spend maps to value instead of hype.
- Visualisation
- Decision funnel from problem fit through privacy, integration, adoption, and measurable outcome.
- Learner output
- An AI buying checklist and a yes/no pilot decision for one candidate tool, with budget cap and exit path.
Expanded from secure local AI and sensitive document themes
AI 101: Use Document AI Safely
Learners classify document risk against four data classes, then write a plain-language allow / ask-first / never-use policy that also picks the right tool tier and leaves an audit trail - the difference between safe document AI and a breach.
- Visualisation
- Document-risk grid across public, internal, confidential, and sensitive data classes.
- Learner output
- A document AI handling policy with allow, ask-first, and never-use categories.
Expanded from microagent and reusable workflow themes
AI 101: Package Repeatable Agent Systems
Learners turn a one-off prompt into a named procedure with a trigger, inputs, standards, steps, output format, examples, and a review rule - then learn exactly which 2026 platform feature (ChatGPT Project, Custom GPT, Claude Project, or Claude Skill) to package it in, and how to hand it off so it survives the handoff.
- Visualisation
- Reusable system stack: prompt, checklist, folder, skill, automation, and plugin.
- Learner output
- A first reusable AI workflow card plus a packaging decision - the specific platform feature (Project, Custom GPT, or Skill) you will build it in and why - ready to test with a teammate.
Expanded from SEO and GEO growth-system themes
AI 101: Grow Visibility in Search and AI Answers
Learners plan three useful topics with buyer questions, citable proof, structured pages, internal links, a clear next step, and an update cadence - engineered so search engines and AI answer engines can understand and quote the business.
- Visualisation
- Visibility signal board covering keywords, questions, proof, schema, freshness, and internal links.
- Learner output
- A three-topic content plan with target buyer questions, citable proof, internal links, a clear next step, and an update cadence - each topic scored for whether an AI answer engine could quote it.
AI Kick Start course - AI basics, 16 core lessons, 6 guided walkthroughs
One complete course, from basics to operational agent workflows.
Start with AI basics and practical AI 101, then continue into 16 workflow lessons and six click-by-click walkthroughs for Codex and Claude. Each walkthrough uses real screens, exact actions, and implementation checkpoints.
Learner outcomes
What you can do by the end.
- Explain when to use Codex, Claude Cowork, Claude Code, desktop agents, CLI agents, skills, plugins, connectors, MCP, and scheduled tasks.
- Operate the Codex App and CLI, Claude Chat, Claude Cowork, Claude Code in the app, and the Claude Code CLI through six built-in screen-by-screen walkthroughs.
- Build a safe personal AI operations folder with instructions, prompt templates, outputs, logs, and review rules.
- Run a practical file-based agent workflow and improve it through a bounded review loop.
- Draft one recurring automation and one mobile-to-desktop task prompt.
- Turn a repeated workflow into a reusable skill and plugin concept.
OpenAI CodexBest for project execution, code, tooling and repeatable technical workflows.Claude Cowork + Claude CodeBest for documents, local files, knowledge work and human-readable output.
Source notes · checked 2026-06-05