DAX Foundation

Meta DAX › Showcase

What it hands a guide and a learner, three steps in.

Three iterations have happened. After each one the guide told Meta DAX how it went, and it learned something about the learner and something about the guide. Each page is what it hands them now: the plan for step four, with steps one to three visible as history. Two big tabs, one for each person, each in its own voice and its own look. Everyone here is made up.

The six

  1. The Six-Out Inning Megan, a mom who homeschools, and Liam, 11

    Adding fractions with different bottom numbers, through baseball. Counted in sessions.

    An inning has six outs, so half an inning is three outs, a third is two, and a half plus a third is five outs: 5/6, not 2/5. Three sessions in, Liam leads the first ten minutes of the fourth, because he knew what 5.2 innings meant before his mom did.

    For Megan
    What you need, how it goes, what to say, what to skip, and the slips that are normal.
    For Liam
    Box Score Boss: a scoreboard game with the decimal trap built in. No AI in it.

    Open the page →

  2. Make Your Case Priya, an English teacher, and Mateus, 15, eight months in Canada

    A persuasive paragraph: claim, evidence, reasoning. Counted in lessons.

    He argues brilliantly out loud and writes two lines. Lesson four is the "so what": the reasoning that links the evidence to the claim. Then he teaches Priya five Portuguese false friends, and she uses his list with her other newcomers.

    For Priya
    One focus per draft, what to say instead of red ink, a mini exit slip.
    For Mateus
    Reasoning Rally and True friend or false friend?, on a futsal court.

    Open the page →

  3. Within Reach Dinah, who teaches teachers, and Jordan, a first-year Grade 3 teacher

    Checking for understanding in the moment; help that fades on purpose. Counted in conversations.

    "Does everyone get it?" Every hand nodded; half the class failed the exit task. Four conversations in, Jordan designs a hinge question with every wrong answer planned for, and teaches Dinah how his students’ families’ work, trap lines and firewood, becomes maths.

    For Dinah Rexford
    Peer to peer, with the real terms: ZPD, contingent support, fading, hinge questions. An observation lens for her next visit.
    For Jordan
    What would you do next? and Fade it, on a chalkboard.

    Open the page →

  4. Everyone in the Room Renée, a people-and-culture lead, and Marcus, a new engineering manager

    Meetings where everyone is heard, across three time zones. Counted in check-ins.

    "Manila never speaks up in standup." The standup sat at nine in the morning in Toronto. Written input now counts as speaking, every meeting ends with decisions written down, and in the next retrospective Joy leads, with the format she ran at her last company. Marcus is the learner in that room and says so.

    For Renée
    How to coach without taking over, and one line on psychological safety.
    For Marcus
    A meeting simulator and a playbook he co-writes with the team.

    Open the page →

  5. First On-Call Dave, a site reliability engineer, and Aisha, on her first on-call rotation

    Handling a live incident; the blameless review. Counted in on-call shifts.

    Her first page alone: checkout errors right after a deploy. She spent twenty-five minutes reading logs for the bug instead of rolling back, and Dave took the terminal. By shift three she rolled back in six minutes and found the bad commit in two, with a tool he had never used. Shift four, she shows him.

    For Dave
    How to coach without grabbing the keyboard; the review he wants from her; what he is learning.
    For Aisha
    Pager Drill, Blame or system?, and a status-update builder.

    Open the page →

  6. Real or Scam? Theo, 19, and Lucille, 78, his grandmother

    Spotting phone, text and email scams. Counted in visits.

    A caller pretending to be Theo in trouble asked for money. She hung up, because he said "Grandma" and Theo calls her Mémère. Visit four: practice, a family code word, who to tell. Then Lucille teaches Theo how she read a stranger across the hardware-store counter for thirty-five years.

    For Theo
    Let her tap, what to set up together, how to praise without talking down.
    For Lucille
    Vrai ou arnaque? / Real or scam?, in large print, French and English.

    Open the page →

What every page has

The guide’s side

  • Up next: the plan for step four, concrete. What you need, how it goes, what to say, what not to say.
  • What we’ve learned: about the learner and about the guide, step by step, each with what changed because of it.
  • The steps: one to three done, four next, five sketched lightly.
  • When it’s tricky: if-then cards, including what to do when the guide feels stuck.
  • Tell us and ask: a pre-filled note to send back after the step, and a prompt the guide can paste into their own AI chat. The guide asks; a child never does.

The learner’s side

  • A completely different look, made for them: a ballpark, a futsal court, a chalkboard, a dashboard, a console, large print in two languages.
  • At least one thing to do that uses no AI: predict or choose first, then see. Every wrong option is mapped to the specific mistake behind it, with its own feedback. It gets harder after a streak and easier after a miss.
  • The questions they might ask next, answered two levels deep.
  • Your turn to lead: the moment the learner is a step ahead of the guide on something, and leads.
  • A learner under eighteen uses their side with the guide beside them. Nothing invites a child to chat with an AI.

How they were made, honestly

Each sample was written with help from AI from a short brief: two living profiles that grow as we learn about each person, what happened in steps one to three, and what step four is for. Every page was read in full by a person before it went up, and fixed where it was wrong: a number that did not add up, a claim of adaptivity the game did not have, a word we do not use. The pages store nothing and send nothing about anyone. They show what Meta DAX is aiming to hand people; the automated pipeline that writes a whole course from one sentence has so far had automated runs only.

The set grows. Profiles are kept and enriched; each scenario is extended one iteration at a time; and the next ones will be built from real scenarios people send us. Send one →