84 courses · 283 lessons
All courses
Each course unlocks independently; memberships unlock all courses.
01 / 03
One-day lessons
Take one useful problem from idea to done today.
52 courses
AI for Everyday Work: Email, Meetings, and Follow-Through
A practical one-day workflow using a fictional workday: triage email, extract meeting actions, and verify a short handoff before anyone acts on it.
AI for Everyday Life: A Realistic Weekly Plan
A practical one-day plan for a fictional household: protect fixed commitments, organize errands and groceries, and verify a weekend outing before spending money.
AI for Proposals: A One-Page Case and a Talk Order
A practical one-day workflow: turn a fictional team's idea into a one-page proposal with checked evidence, honest options, and a five-slide talk order for the decision-maker.
AI for Reading Long Documents: Quote, Check, and Ask
A practical one-day workflow: use a fictional lease to make AI quote clauses before it answers, check every quote against the source, and list missing documents and questions to ask.
AI for Bills and Subscriptions: Sort, Check, and Compare
A practical one-day workflow: use a fictional month of transactions to have AI categorise spending, flag subscription charges that need checking, compare phone plans on their full terms, and recompute every figure yourself.
AI for Job Applications: Tailor Honestly, Then Rehearse
A practical one-day workflow: use a fictional candidate and job post to map requirements to evidence, rewrite resume bullets without inventing anything, and run a mock interview with AI scored against a rubric.
AI for Parent-Led Homework Help: Keep the Child Thinking
A one-day parent-led lesson using nine-year-old Mina's fraction question: ask for hints, verify explanations, and leave the child's answer to the child.
AI for Family Routines: School, Homework, and Rest
Plan a realistic school week for nine-year-old Mina, protecting sleep, play, travel buffers, and the child's say in the plan.
AI for Kids' Active Play: Plan Movement They Enjoy
Build an adjustable active-play menu for nine-year-old Mina using AI for ideas while checking public-health guidance and keeping health decisions with adults.
AI for Family Meals: Plan Together, Verify Before Serving
Plan two family dinners and two lunchboxes for nine-year-old Mina, using AI for options and a caregiver for school-rule, ingredient, and allergy checks.
AI for Everyday Planning and Travel
Practice organizing information and drafting realistic plans while checking travel details and budgets.
Getting Started with Doubao, DeepSeek and Tencent Yuanbao
Find trusted official entry points and practice asking, revising and checking an answer.
AI Poster Tutorial: From Event Brief to Text Review
Plan a fictional event poster, generate a background, add text and check it before sharing.
How to Write AI Image Prompts: Subject, Setting and Revision
Plan a fictional reading-event illustration, write a focused image prompt, and revise it using observable results.
Before Uploading Photos to AI: Check Other People’s Privacy
Use a fictional desk photo to review people, background details, file information and service terms before deciding whether to upload.
AI Images for WeChat Moments: Size, Content and Labels
Use a fictional book-sorting post to choose an image, review its preview and claims, and decide whether it is ready to share.
Write a Spoken Script with AI: Timing and Delivery
Use a fictional event to choose a speaking topic, supply facts, time a natural read and revise the script.
AI Product Images: Prepare Assets and a Clear Brief
Use a fictional mug to organize product facts, asset permissions and an image brief, then check the image and copy before use.
Crop and Adjust Phone Photos: From Purpose to Review
Use a fictional desk photo to plan a crop, inspect the subject and decide whether AI editing is needed.
Choosing AI Voice and Captions: Check Pronunciation and Rights
Use a fictional reading-event clip to choose narration, review caption timing, and check permission and privacy.
Can You Use AI Images Commercially? Permissions to Check First
Use a fictional craft-shop promotion to check tool terms, source permissions and the intended use, then record the evidence gaps.
How to Request AI Image Revisions: Observe and Compare
Use a fictional reading illustration to observe an existing image, request specific changes, compare versions and review usage conditions.
Before Publishing an AI Video: Check Facts, Rights, and Labels
Review a fictional clip shot by shot for meaning, facts, privacy, permissions, labels, and publication preview.
AI Customer Replies: Common Questions and Human Handoffs
Use a fictional shop fact card to practice FAQ replies, exception routing, privacy and review before sending.
AI Product Copy: Facts, Selling Points and Limits
Use fictional notebook information to draft product copy for different channels and check each claim against facts, uses and missing evidence.
Practice Speaking with AI: Ask for Specific Feedback
Start with your own fictional speaking practice, request checkable feedback, decide what to use, and practice again.
Check AI Spreadsheet Formulas with a Small Sample
Use a fictional product table to inspect prices, quantities, formula ranges and totals before trusting an AI draft.
AI Event Copy: Headlines, Audience and Fact Checks
Use a fictional shop event to identify the audience, facts and open questions, then draft and check copy for different channels.
Before AI Organizes a Table: Fields and Privacy
Use a fictional supply log to define the task, inspect fields and missing values, and prepare safe inputs and acceptance checks.
How to Check AI Explanations: Examples, Sources, and Counterexamples
Use a reproducible coin problem to check an explanation's definition, examples, conditions, and citation.
AI Table Cleanup: Find Duplicates Without Losing Records
Use fictional order lines to identify duplicate candidates, missing values and conflicts, then compare retained row counts and amounts after human review.
AI for Sharing Housework: Adjust a List Together
Use a fictional household to list chores, discuss choices and rotation, and revise an unconfirmed plan using feedback.
Write a Self-Introduction with AI: Real Experience and Clear Structure
Use a fictional applicant to select relevant experience, check each sentence against evidence, and rehearse a self-introduction.
AI Feedback Sorting: Classify Without Changing the Meaning
Use fictional shop feedback to practice item-by-item classification, multiple topic labels and separation of original wording from inference, with scope and open questions recorded.
Compare Products with AI: Specifications, Total Price and Returns
Use two fictional storage boxes to compare specifications, recalculate item and shipping costs, and record price and return questions.
Read an Article with AI: Separate Observations, Opinions, and Unknowns
Use a fictional science article to distinguish reported observations, author views, sample limits, and missing information.
Polish Notices with AI: Preserve Conditions and Open Questions
Compare a fictional original notice with a revised draft while preserving dates, audience, conditions and unconfirmed details.
Scale a Recipe with AI: Proportions, Units and Checks
Use a fictional four-serving recipe to scale to six servings, check units and a fractional ingredient, and record questions needing separate review.
Read Charts with AI: Axes, Units, and Change
Use fictional monthly data and a text bar diagram to read axes, handle a missing value, and check changes.
Design a Survey with AI: Clear Questions, Neutral Choices
Draft a short survey for a fictional reading event, practice neutral questions and clear options, then revise through paper walkthroughs.
Use AI to Make a Packing List: Duration, Context and Open Questions
Use a fictional two-day, one-night trip to choose items, check quantities and shared arrangements, and prepare a return checklist.
Translate a Short Notice with AI: Check Dates, Conditions, and Meaning
Use a fictional notice to compare Chinese and English sentence by sentence, checking dates, negatives, conditions, and unresolved details.
AI for After-Sales Requests: State Facts and What You Need
Use a fictional product issue to organize observations, evidence and open questions, then draft an unsent consumer request.
After AI Reads Image Text: Check Numbers, Punctuation and Omissions
Use a fictional text layout and a simulated recognition draft to review characters, dates, amounts, punctuation and missing lines.
Prepare a Short Meeting with AI: Goals, Topics, and Time
Use a fictional 30-minute discussion to select pre-meeting topics, allocate time, and retain open questions.
Use AI to Describe Images: Explain the Scene Without Guessing Identity
Use a fictional scene to write short text alternatives for different contexts, keeping verifiable details and avoiding guesses about people.
Turn Instructions into a Checklist: Order, Conditions and Open Items
Use fictional library instructions to preserve sequence, applicable conditions and unknowns while distinguishing action status.
Compare Two Versions with AI: Changes and Open Questions
Use two fictional rule drafts to review additions, removed text, changed meaning, unchanged content, and unknowns.
Use AI to Prepare Service Questions: Describe, Then Ask
Use a fictional old desk lamp to organize observations and service questions without guessing a cause, booking or sending.
Use AI to Organize File Names: Dates, Versions and Handoffs
Use five fictional files to distinguish creation dates, content dates and version status, and make naming suggestions for review.
Before Giving AI Materials: Decide What to Leave Out
Use a fictional event schedule to choose necessary information, create a separate example, and record sharing status.
Pause Before a Link: Use AI to Plan Independent Checks
Compare reserved example-domain text on paper, notice spelling and hierarchy differences, and plan independent verification without visiting links.
02 / 03
Focused courses
Build one skill over about a week.
28 courses
AI for Beginners: Start on Your Phone
Start a conversation on your phone, ask a clear question, revise the answer, and check the details without coding.
Prompt Engineering From Scratch in 5 Days
For anyone who has used an AI chat product: in five days, learn to turn a fuzzy requirement into a prompt the model executes reliably, and verify it with a small test set.
Mastering Claude: From Conversation to Claude Code in 5 Days
In five days go from chatting with Claude to running a real development task end to end with Claude Code — explore, plan, implement, verify, commit — then wire up the API and the Agent SDK.
Mastering Codex and the OpenAI Agents SDK in 5 Days
In five days get fluent with the Codex CLI, cloud tasks, the Responses API, and the Agents SDK, then spend the last day running the same task through both Codex and Claude Code for a real side-by-side comparison.
MCP in 7 Days: Wire Tools Into Any Agent
For developers who write TypeScript or Python: in seven days, learn to read, implement, debug, and safely ship MCP servers and clients against the 2026-07-28 spec.
Agent Skills in 7 Days: Turn Experience Into Reusable Capability
In seven days, learn to turn something you already do well into a SKILL.md, so any client that supports Agent Skills opens it up on its own when it's needed, and package a team's conventions for distribution.
Context Engineering in 5 Days
Treat the context window as a piece of luggage with fixed capacity: in five days, learn to break down what fills it, work out its bill, and keep it under control with pruning, compression, and isolation.
Frontend Agent UX in 7 Days: From Streaming to Generative UI
For React developers: in seven days, hand-build an agent workbench in the browser. Align with the open AG-UI protocol and turn a raw event stream into an interface that renders without jank, pauses for approval, replays history, and reads correctly to a screen reader.
Agent Evals and Observability in 7 Days: Turning "Seems Fine" Into Numbers You Can Reproduce
For anyone who has shipped an agent: build a zero-dependency evaluation framework in seven days. Turn non-determinism into pass^k, mine your incidents for a benchmark set, put the model judge itself through calibration, score the trajectory rather than just the ending, wire up telemetry and a cost panel, and finally block regressions before they merge.
Agent Security in 5 Days
Take one internal assistant that happens to hold all three parts of the lethal trifecta and turn it into a target: five days from threat modeling to architectural defense, runtime controls and a red-team loop that drives attack success to zero while keeping the agent useful.
Agent Harness in 7 Days: Let an Agent Run Unattended Through the Night
For anyone who has already shipped at least one agent: hand-build an unattended runtime in seven days. Move the authoritative state out of the context window, keep progress alive on disk and in git, block premature completion claims with an end-to-end gate, intervene automatically on doom loops and blown budgets, and finally let it run through the night and hand you a report you can actually check in the morning.
Multi-Agent Systems and the A2A Protocol in 7 Days: From In-Process Orchestration to Cross-Team Collaboration
For anyone who has built at least one agent. First, reproducible experiments answer when a task should be split across several agents and where information leaks at every handoff. Then you hand-build an A2A v1.0 server and client and treat another team's black-box agent as a teammate: discover it by its card, track it through tasks, trust it through signatures, and still deliver an honest result when something breaks.
Agent Engineer Interviews in 7 Days: Round by Round, Still Standing at the Third Follow-Up
For anyone who has taken at least one agent course and is getting ready to apply. Day by day, it breaks the interview into its rounds, concept questions, project deep dives, system design, incident scenarios and coding, and behavioral questions, and drills every answer until you can say why you did not do it the other way. Along the way you hand-build an interviewer agent that asks follow-ups, scores answers, and catches fabricated claims.
ChatGPT for Real Work: From a Vague Request to a Verified Deliverable
Five practical days to turn a messy request into an evidence-backed result. Brief the task, analyze files, research claims, reuse context in a project, and verify the final handoff.
Gemini and Google AI Studio in 5 Days: From Prompt Experiments to a Verified Handoff
Five days with fictional bilingual support requests: test prompts and multimodal inputs in AI Studio, design structured output, separate grounding from actions, and hand off a reproducible workflow.
AI Automation With n8n in 5 Days: From One Run to Reliable Triage
Build a five-day workflow for fictional bilingual support tickets: triggers, data retrieval, mappings and branches, AI classification, then errors, duplicates, and human approval.
Gemini Notebook (formerly NotebookLM) in 5 Days: From Source Selection to an Auditable Handoff
Use a wholly fictional urban shade study to select permissible sources, verify citations, resolve conflicting evidence, create briefing and study outputs, and hand off a reviewable research package.
Cursor Agent Workflow in 5 Days: From Codebase Exploration to Cloud Handoff
Use a fictional repair-booking repository to practice scoped exploration, small Agent edits, rule scope, multi-file review, and safe Cloud Agent handoff. English is the source text; Chinese is its faithful translation.
Gemini in Google Sheets in 5 Days: From Input Audit to Reproducible Handoff
Use fully fictional community event registrations and budgets to audit inputs, review formulas, reconcile charts, detect errors and bias, and hand off a reproducible analysis. English is the source manuscript; Chinese is a faithful translation.
Perplexity Research in 5 Days: From Search Questions to an Auditable Handoff
Follow a wholly fictional Harborfield night-bus proposal to discover and cross-check web sources with Perplexity and Pro Search, handle freshness and conflicts, write a bounded brief, and hand off reviewable evidence.
Adobe Firefly Creative Assets in 5 Days: From Brief to Delivery Review
Create an image-led marketing kit and a short video concept for a wholly fictional community arts weekend. The English course is the source manuscript; its Chinese edition is a faithful translation.
AI Study Coach in 5 Days: Learn It Yourself, With AI's Help
Follow a fictional learner, Noor, through two weeks of self-study in introductory statistics: set checkable goals and take a diagnostic, explain and verify, test yourself, space your review, and respect assessment rules.
Practise a Language With AI in 5 Days: Role-Play, Writing Feedback, Review
Follow a fictional learner, Lin, preparing for a three-day work trip: set scenario goals and a practice-partner brief, role-play with AI, get writing feedback by error type, check phrase cards against a dictionary, and finish with a full simulation and self-assessment. This English edition practises Spanish; the Chinese edition practises English with the same structure.
Scam-Proof in the AI Era: Spot Fakes and Verify Before You Pay
Use fictional scenarios to recognise cloned voices, phishing messages, and fake support, verify independently, make a family plan, and respond after a scam.
5 Days of AI Office Work and Spreadsheets: From Notes to a Chart-Based Briefing
Practice meeting notes, weekly updates, table checks, formulas and a three-slide briefing with a fictional community event.
AI Writing: Drafts, Editing and Email
Use fictional examples to draft, revise and check writing without losing your meaning.
AI Image Making in 5 Days: From a First Picture to a Practical Poster
Start with an illustration on a phone or computer, then practice prompting, revision, layout and checks before use.
Plan a Short Video Script with AI: Hook and Shot List
Turn a fictional reading-event idea into a shootable opening, shot list and review checklist.
03 / 03
Long-form projects
Follow a longer path and ship something real.
4 courses
From Frontend Engineer to Agent Engineer in 30 Days
For anyone with no AI background: four weeks from the LLM API all the way to multi-agent orchestration and a job-hunt sprint.
RAG in 14 Days: From Retrieval to Trustworthy Answers
For developers who write TypeScript or Python: in fourteen days, turn Retrieval-Augmented Generation from a working prototype into an evaluable, attributable, production-ready service — get keyword search solid first, then push it layer by layer toward production with vectors, reranking, query rewriting, and evaluation.
Build an AI Short-Drama Production Pipeline With Agents in 14 Days
For developers who write TypeScript: in fourteen days, turn a single sentence into one episode of a vertical short drama, then scale it into a production platform that can produce in parallel, be reviewed, and keep cost under control.
Build Your Own Coding Agent in 21 Days
For developers who write TypeScript: in twenty-one days, start from an empty-shell REPL, add one capability a day, and end up with a terminal Coding Agent that can read code, edit files, run tests, and work in parallel on a real repo, installable and distributable, with zero framework and zero runtime dependencies throughout.