🇲🇾 Your 3-Month Coursera Roadmap
For Malaysians using the Rakyat Digital Global Learning Pass
You have 16,000+ courses and only three months. You do not need more options. You need a good first move, then one direction.
This is a starting map, not a checklist. Finish a course, make one small thing from it, then choose what to learn next.
1. Start here: a strong general AI foundation
Default starting point: Google AI Professional Certificate
If you do not know what to choose, start with the Google AI Professional Certificate.
It is the clearest default because it gives you a structured, beginner-friendly foundation across the everyday places AI shows up at work: planning, research, writing, content, data and building. You will finish with more than a general idea of AI. You will have tried it on real tasks.
By: Google
Estimated: 8 hours in total, self-paced
What you will learn: Practical prompting and responsible AI, then applied work in planning, research, communication, content creation, data analysis, app building and app deployment.
Best for: Beginners who want one structured route instead of picking random courses
Tiny project after it: Build one small AI helper for a repeated problem in your work, business or studies. Test it with three real examples and write down where you still need human judgement.
It currently includes these eight courses:
- AI Fundamentals
- AI for Brainstorming and Planning
- AI for Research and Insights
- AI for Writing and Communicating
- AI for Content Creation
- AI for Data Analysis
- AI for App Building
- AI for App Deployment
Want to understand how AI actually works?
Choose Introduction to Artificial Intelligence (AI) by IBM instead of, or before, the Google certificate if you want more of the underlying ideas.
By: IBM
Estimated: about 10 hours
What you will learn: AI, machine learning, deep learning, neural networks, generative AI, common applications, ethics and governance. It includes hands-on labs and a final workplace-use project.
Best for: Curious beginners who want to understand the ideas behind AI, not only use workplace tools.
Tiny project after it: Make a one-page map of one real process around you. Mark where AI could help, where it could fail, and who should still make the final call.
2. Then choose your learning path
Do the Google certificate first if you are new. Then choose one path based on what you want to make or improve. The Google courses appear again below where they fit, because you may want to revisit them with a specific goal.
🤖 AI Agents & Automation
AI Agents 101: Foundations of AI Agents
By: LearnQuest
Estimated: 4 hours
What you will learn: How agents differ from chatbots and ordinary automation, plus tools, memory, workflow mapping, risk and governance.
Best for: Non-technical people who need a sensible mental model before automating work.
Tiny project after it: Map one repetitive task. Write the trigger, actions, information needed, human approval point and what could go wrong.
ChatGPT & Zapier: Agentic AI for Everyone
By: Vanderbilt University
Estimated: 8 hours
What you will learn: Build practical agents with Custom GPTs and Zapier that can act in tools such as Gmail and Google Sheets.
Best for: Beginners who want to make one useful, low-risk automation without becoming a programmer.
Tiny project after it: Build a draft-only automation for form responses, ideas or receipts. Test it with fake data before connecting real accounts.
Agentic AI
By: DeepLearning.AI
Estimated: about 20 hours
What you will learn: Reflection, tool use, planning and multi-agent patterns, with Python, evaluation and production testing.
Best for: Technical learners who can already work comfortably with Python and want to build reliable agentic systems.
Tiny project after it: Build a tiny research workflow that plans, gathers and critiques a draft. Keep a human approval step before anything is used.
📊 Data + AI
AI for Data Analysis
By: Google. This is course six in the Google certificate.
Estimated: 1 hour
What you will learn: Define success metrics, clean messy data, create Google Sheets formulas with natural-language prompts and make charts.
Best for: Anyone who works with spreadsheets, reports or basic business data.
Tiny project after it: Clean one small spreadsheet, create one chart and write three findings. Check every number yourself.
AI-Enabled Data Analytics
By: Chegg Skills
Estimated: about 20 hours
What you will learn: Use AI for prompting, data preparation, trend-finding, visualisation, business decisions and responsible analytics.
Best for: Beginners who want a slower, fuller data workflow and do not need programming.
Tiny project after it: Use a public dataset to answer one question. Make one chart, list three findings and record what you manually checked.
GenAI for Data & Analytics
By: STARWEAVER
Estimated: 9 hours
What you will learn: Use ChatGPT, Gemini, Python and Google Colab across the analytics lifecycle, from problem framing to data storytelling.
Best for: Analysts and business users with some data confidence who want to go beyond spreadsheets.
Tiny project after it: Turn a vague business question into an analysis brief, one visual and a short decision memo. Ask AI to critique the logic, then fact-check it.
🎨 Content Creation
AI for Content Creation
By: Google. This is course five in the Google certificate.
Estimated: 2 hours
What you will learn: Generate and critique images and video, create presentations and set simple design guidelines with Gemini.
Best for: Creators, marketers and anyone making visual work.
Tiny project after it: Turn one content idea into a visual brief, two asset variations and a small checklist for reviewing the final result.
AI Content Foundations — Tools, Prompting, and Brand Voice
By: LearnQuest
Estimated: 3 hours
What you will learn: Choose AI tools for a job, write stronger prompts, encode brand voice, direct visuals and make a repeatable content workflow.
Best for: Creators and marketers who want AI output to sound and look more like them.
Tiny project after it: Turn one old post or product page into a repeatable workflow with a voice guide and a final quality check.
📣 Marketing + AI
Artificial Intelligence for Marketing
By: Madecraft
Estimated: 3 hours
What you will learn: Where AI fits in marketing, including personalisation, advertising, customer engagement and the basics behind the tools.
Best for: Marketers who want a short strategic overview before changing their day-to-day workflow.
Tiny project after it: Pick one campaign and make an AI opportunity map: research, ideas, production, targeting, reporting and human review.
Digital Marketing with AI
By: SkillsBooster Academy
Estimated: 5 hours
What you will learn: AI-assisted content strategy, customer segmentation, campaign personalisation, creative work and reusable marketing workflows.
Best for: Marketers who want to apply AI across a real campaign.
Tiny project after it: Create one campaign brief with a customer segment, three content angles, one email and a review checklist.
AI for E-Commerce: No-Code Tools for Sales & Marketing
By: Board Infinity
Estimated: 4 hours
What you will learn: Product content, customer-service chatbots, product recommendations, customer segments, demand forecasting and pricing using no-code tools.
Best for: Online-store owners and e-commerce marketers.
Tiny project after it: Improve one product listing, draft five customer-service replies and create a simple review-insight sheet.
💼 Business + AI
Negotiation Skills Using AI
By: Chris Croft Training
Estimated: about 10 hours
What you will learn: A step-by-step negotiation process, with AI for research, scenarios, questions, tactics and decision support.
Best for: Founders, managers, salespeople and anyone who negotiates with clients, suppliers or teams.
Tiny project after it: Prepare for one real or imaginary negotiation: your goal, walk-away point, tradeables, questions and three possible outcomes.
🧠 Critical Thinking & Decision Making
Critical Thinking for Better Decisions in the ChatGPT Era
By: Deep Teaching Solutions
Estimated: 3 hours
What you will learn: How bias, emotions and fast thinking affect decisions, including how to question AI-generated information.
Best for: Everyone. This is the strongest short course here for using AI without switching off your judgement.
Tiny project after it: Ask AI a question you care about. Verify three claims with real sources, then write what changed after checking.
Critical Thinking: The Human Skill AI Can't Replace
By: STARWEAVER
Estimated: 5 hours
What you will learn: Citation audits, bias checks, risk reviews, decision journals and defensible decision memos for AI-assisted work.
Best for: Professionals and managers who need a practical framework for deciding whether an AI answer is safe to use.
Tiny project after it: Take one AI-generated recommendation and write a one-page decision memo with sources, risks, assumptions and your final call.
💻 Vibe Coding / Building with AI
AI for App Building
By: Google. This is course seven in the Google certificate.
Estimated: 2 hours
What you will learn: Find a workflow worth improving, turn a plain-language brief into a working app and use Google AI Studio responsibly.
Best for: Beginners who want to test whether building with AI is for them.
Tiny project after it: Build a small calculator, checklist or tracker for one annoying task. Test five actions before sharing it.
Vibe Coding: AI-Powered App Development for Product Managers
By: University of Maryland, College Park
Estimated: 9 hours
What you will learn: Natural-language app design, context engineering, Model Context Protocol, testing and AI-assisted product cycles.
Best for: Founders, creators and product-minded beginners who want to make a useful first prototype.
Tiny project after it: Write a one-page app brief, make a prototype and ask three people to try one task without your help.
Building Full-Stack Applications with Vibe Coding
By: Edureka
Estimated: 8 hours
What you will learn: Build, debug and deploy fuller apps with Bolt, Replit, Lovable, databases and authentication.
Best for: Learners who have made a first prototype and are ready for more moving parts.
Tiny project after it: Turn your prototype into a tiny working app with one user flow, simple data storage and a clear error message.
Claude Code for Vibe Coding
By: Edureka
Estimated: 7 hours
What you will learn: Claude Code, terminal-based agents, MCP, task delegation, testing, debugging, security and deployment.
Best for: People with basic programming knowledge who want to work with AI inside a real development workflow.
Tiny project after it: Use Claude Code on a small existing project to fix one bug, add one small feature and write tests that prove it works.
High-signal depth boosters
These are prominent, well-reviewed courses that deserve a place here. They overlap with the foundation or specialist paths above, so use them as choices, not extra homework.
Choose one: a wider generative-AI perspective
Generative AI for Everyone
By: DeepLearning.AI
Estimated: 6 hours
What you will learn: What generative AI can and cannot do, common use cases, prompting, project lifecycle, business opportunities and risks.
Best for: Someone who wants a concise, non-technical perspective on generative AI from Andrew Ng.
Tiny project after it: Write a one-page opportunity and risk brief for one task in your work or business.
Generative AI: Introduction and Applications
By: IBM
Estimated: 8 hours
What you will learn: How generative AI differs from other AI, plus its use across text, image, audio, video and code. It includes tool labs and a final multi-format project.
Best for: Someone who wants a broader tour of generative-AI media and tools, with more hands-on practice than the DeepLearning.AI overview.
Tiny project after it: Use one brief to make a text, image and code or spreadsheet output. Compare where each output helped and where it failed.
[!tip] Which one should I choose? Choose Generative AI for Everyone for the clearest big-picture view. Choose IBM’s Generative AI: Introduction and Applications if you want to try more formats and tools. You do not need both before moving on.
Choose one: get much better at prompting
Google Prompting Essentials
By: Google
Estimated: 4 hours
What you will learn: A five-step prompting framework, iteration methods, multimodal prompts and reusable prompts for work, research, data and presentations.
Best for: Beginners who want a practical prompt library quickly.
Tiny project after it: Build five reusable prompts for a repeated task. Save the best version, one example output and the check you use before trusting it.
Prompt Engineering for ChatGPT
By: Vanderbilt University
Estimated: about 20 hours
What you will learn: Prompt patterns, advanced prompting approaches and how to make prompt-based applications for work, business or education.
Best for: Someone who already knows the basics and wants a more thorough, ChatGPT-focused course.
Tiny project after it: Turn a multi-step task into a prompt-based mini app or workflow. Add a fact-check step and test it with three realistic inputs.
[!tip] Which one should I choose? Choose Google Prompting Essentials for a fast, broad beginner route. Choose Vanderbilt when prompting is the skill you specifically want to practise in depth. The Google AI Professional Certificate already covers prompt basics, so take one of these only if you want more repetition and range.
Technical depth: how LLMs work
Generative AI with Large Language Models
By: DeepLearning.AI and AWS
Estimated: about 20 hours
What you will learn: LLM lifecycle, transformers, fine-tuning, evaluation, deployment and the trade-offs behind real-world LLM applications.
Best for: Technical learners with Python and basic machine-learning knowledge. It is not a beginner replacement for the Google certificate.
Tiny project after it: Write a model-selection and evaluation plan for one LLM use case. Define the data, success measure, failure cases and human review process.
A simple way to use the three months
- Weeks 1 to 2: Finish the Google certificate, or IBM if you want the deeper foundation first.
- Weeks 3 to 7: Pick one path and complete one main course. Make the tiny project.
- Weeks 8 to 12: Take one follow-up course only if it helps you improve that same project. Share what you made, what worked and what you would change.
The goal is not to collect certificates. It is to leave the pass with a stronger judgement, one useful thing you made and a clearer idea of what you want to learn next.
Made by @chloesinyin 💛