Claude 101
The friendliest way into Claude for everyday work. Covers writing, analysis, summarising and managing conversations.
Eight short questions about your goal, your time and your background, plus a quick check of what you already know. Then we build an ordered path of free courses that fits your level.
About two minutes. No account needed, and your answers stay on your device.
The friendliest way into Claude for everyday work. Covers writing, analysis, summarising and managing conversations.
Google's entry course on using AI at work: writing prompts and speeding up routine tasks. Not fully free, so check the cost before enrolling.
Andrew Ng's course for non-coders: what generative AI can and can't do and how to apply it in your own job.
OpenAI's learning hub with workshops, videos and a learner community. A good starting point for ChatGPT in practice.
AI in manager language, not programmer language: what models and data mean and how to lead a project that works.
A fully free University of Helsinki course explaining AI basics without heavy maths. Solid grounding before you touch tools.
A short Google path (about 2 hours) on generative models and responsible use, with a free badge on completion.
A structured framework for working with AI effectively, safely and ethically: when to delegate, when to review, how to judge output.
What models do well, where they fail and why they sometimes hallucinate. Builds calibrated trust instead of blind faith or fear.
A version for small-business owners: using AI in marketing, customer support and operations without a tech team.
Hand long knowledge tasks (research, analysis, documents) to Claude and review the result. Built for non-coders who want to automate work.
Designing work shared between people and AI agents: roles, review and accountability. Useful for team leads.
Andrew Ng on agentic workflows: reflection, tool use and planning. Videos are free; labs and certificate need a Pro plan.
A free course from idea to working agent, with hands-on projects and a free certificate. One of the best open resources on agents.
A 15-lesson open-source curriculum on agent design patterns, tools and memory, with runnable code.
About an hour on LangChain essentials: models, tools and your first agent. A good first step before LangGraph.
A short introduction to building stateful agent graphs and control loops with LangGraph.
The deeper LangGraph course (about 6 hours): memory, human-in-the-loop and deployment. For developers building production agents.
Understand MCP and how it connects AI to your tools and data, then build a simple server and client.
The advanced MCP follow-up: transports, security and scaling. Take the introduction first.
How to split complex work across specialised subagents instead of one long conversation. Short and direct.
Learn to write Skills: reusable instructions that teach Claude your way of getting work done.
A free course, made with Anthropic, on building and shipping MCP applications with hands-on examples.
About 7 hours on prompting fundamentals for anyone, with templates you can apply straight to your own work.
A library of short courses (about an hour each) on RAG, evaluation, reasoning and more. Videos are free.
Twelve free chapters on LLMs: Transformers, fine-tuning and inference. An excellent hands-on reference for developers.
Twenty-one open-source lessons for building generative apps: prompts, RAG and agents, with multi-language code.
The most complete path to building on the Claude API: tools, caching, evaluation. About 9 hours.
A quick tour of the developer platform: keys, models and your first call. Start here before the full path.
The API path delivered through Amazon Bedrock, useful if your company runs on AWS. About 8 hours.
The same path through Google Cloud's Vertex AI for teams on Google Cloud. About 8.5 hours.
Thirty short lessons (about 1.5 hours) on building LLM apps with LangChain: chains, tools and retrieval.
Build your first app by describing it instead of hand-coding, taught by the Replit team. About 1.5 hours; fine for non-coders.
Two hours with an Anthropic educator on driving Claude Code in real projects. Free to enter at the moment.
Your first steps with Claude Code in the terminal: installing, understanding files and finishing small coding tasks.
One hour of practice: fitting Claude Code into daily work (tests, review, refactoring).
A version for product builders: how to think and collaborate with AI while designing and developing.
A free introduction to describing software to AI tools, suited to complete beginners.
A Samara Stack article comparing free and open-source AI coding tools and how to choose.
A free top-down course: build first, then dig into theory. One of the best routes to a practitioner level in deep learning.
Eight videos (about 12 hours) building neural nets and a GPT-style model from scratch. For real understanding.
Short free interactive notebook courses in Python, machine learning and data. Great for fast practice.
A large free curriculum (about 300 hours per the platform) on machine learning with Python, with certification projects.
MIT's open lectures and labs on deep learning, refreshed every year. For those with some programming background.
Harvard's open course on AI concepts in Python: search, knowledge, learning. Free to watch; the verified certificate is paid.
MIT's undergraduate machine-learning course with open materials. Best if you are comfortable with maths.
Stanford's well-known machine-learning lecture series. Deep, and it expects linear algebra and probability.
Google's free, fast-paced machine-learning course with interactive exercises. A quick gateway to the core ideas.
A 12-week, 24-lesson MIT-licensed curriculum covering symbolic AI, neural networks and language processing.
Twenty-six lessons and 52 quizzes of classic machine learning over 12 weeks, MIT licensed.
Free courses on running machine learning on tiny devices and sensors. A good entry to edge AI.
Hands-on training and deployment of a model on a microcontroller using Edge Impulse, whose developer plan is free.
A step-by-step Arabic guide to running a model on your own machine with no account or API key, including hardware needs.
Meet the open Falcon-H1 models from Abu Dhabi's Technology Innovation Institute and how to try them locally. Start here for Gulf-built models.
One hour from PMI on generative AI in project management; earns 5 PDUs with no membership needed.
Five practical hours on using AI for planning, risk and reporting in your projects.
A free introduction to running AI projects (CPMAI): the project lifecycle, and when AI projects succeed or fail.
A page grouping seven free PMI courses, including prompt engineering for PMs and agentic AI for PMs. Some are members-only.
Your suggested path
Understand AI well enough to steer decisions and work with it
Your suggested path
Agents and automation: finish in minutes what took days
Your suggested path
Prompts and LLMs: turn words into results
Your suggested path
Ship your first app by describing it, with AI doing the typing
Your suggested path
Deep theory: how neural networks actually work
Your suggested path
AI on your own machine and at the edge: local, open, no subscription
Your suggested path
A regional track for project managers and engineers: managing with AI