← Open coding lab

NORTHRIDGE · ORIGINAL AUTHORING DRAFT · OCTOBER 30, 2026

WorldForge Lab

A teen creator course with Smart AI One. Sixteen sessions move from controlled experiments to an original playable challenge. This is an independently authored draft, pending a full review of the partner curriculum.

Prerequisites: keyboard and browser basics; no coding experience. Tools: this browser lab; optional Roblox Studio for the transfer extension. Session rhythm: 5 minutes predict, 10 model/trace, 10 edit/run, 15 play/create, 10 explain and reflect. Learners choose themes and difficulty goals.

Coaching: ask for predictions and evidence before offering a fix. Offer a hint, a smaller test, then a worked example only when needed. In the prototype Smart AI One is scripted coaching, not a connected model. Encourage ownership; avoid ranking learners by speed.

UNIT 01 · TWO 50-MINUTE SESSIONS

Movement as a design choice

Concepts. Variables, numeric values, assignment and cause/effect.

Session A — investigate. Measure the time to move between two landmarks at speed 6, then 12. Predict the ratio before running.

Session B — create. Tune speed and jump height for a relaxed route and a precision route.

Evidence of learning. Explain what stayed constant and why changing two variables would weaken the comparison.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 02 · TWO 50-MINUTE SESSIONS

A world made of objects

Concepts. Coordinates, dimensions, composition and spatial reasoning.

Session A — investigate. Draw a route on paper. Translate each ledge into platform(x,y,width), run after each addition, and compare the sketch with the result.

Session B — create. Build a readable path with an optional harder branch and three reachable coins.

Evidence of learning. A peer reaches the goal without spoken guidance; the creator explains one accessibility improvement.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 03 · TWO 50-MINUTE SESSIONS

Rules and difficulty

Concepts. Comparisons, conditionals and mutually exclusive branches.

Session A — investigate. Trace both branches by hand before changing difficulty. Test values below, at and above the threshold.

Session B — create. Create two difficulty modes through jump and speed settings.

Evidence of learning. Show three boundary cases and identify which branch runs in each.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 04 · TWO 50-MINUTE SESSIONS

Patterns and repetition

Concepts. Bounded loops, counters and arithmetic expressions.

Session A — investigate. Predict all coordinates for i = 1, 4; run and compare each object. Change spacing without changing the repetition count.

Session B — create. Build a staircase with six repeated elements and a deliberate variation.

Evidence of learning. Explain the difference between iteration count and spacing; repair an off-by-one mistake.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 05 · TWO 50-MINUTE SESSIONS

Reusable systems

Concepts. Function definitions, calls and decomposition.

Session A — investigate. Define a zero-argument route function, predict what happens before calling it, then call and inspect.

Session B — create. Separate route creation and coin creation into named functions.

Evidence of learning. Explain why definition alone creates no objects and describe a useful function name.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 06 · TWO 50-MINUTE SESSIONS

A world that reacts

Concepts. Events, persistent state and score.

Session A — investigate. Predict score after two coin contacts. Define onCoin, play to trigger it, and compare with prediction.

Session B — create. Award 10 points per coin and add a threshold bonus with a condition.

Evidence of learning. Distinguish setup execution from event execution and demonstrate score persists across contacts.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 07 · TWO 50-MINUTE SESSIONS

Debug with evidence

Concepts. Syntax, runtime and design bugs; controlled experiments.

Session A — investigate. Reproduce a route that cannot be jumped. Write expected/observed behavior, isolate one value, and retest. Then test an unknown variable error.

Session B — create. Repair a partner’s intentionally flawed route with the smallest justified changes.

Evidence of learning. Submit a four-part bug report and a reproducible verification path.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

UNIT 08 · TWO 50-MINUTE SESSIONS

Creator release

Concepts. Systems integration, playtesting, iteration and communication.

Session A — investigate. Pitch a 60-second challenge, sketch the route and write acceptance criteria before coding.

Session B — create. Combine a variable, condition, loop, function and event; run two peer tests and revise.

Evidence of learning. Deliver saved source, test notes, a creator statement and a Studio movement starter. Grade reasoning and iteration as well as playability.

Smart AI One prompt. “What do you expect, what did you observe, and which single change would help you learn more?”

Assessment and transfer

Use four dimensions scored 0–3: behavior matches intent, code reasoning, evidence from tests, and iteration/communication. 0 = absent; 1 = partial with support; 2 = independent and correct; 3 = explains tradeoffs and transfers the idea. Reflection completion alone does not establish mastery.

Optional Studio extension: open a personal blank place, add a LocalScript in StarterPlayer → StarterCharacterScripts, paste the exported movement starter and test a character. Explain the mapping between Forge speed/jumpHeight and Humanoid WalkSpeed/JumpHeight. Build platform objects manually and compare the coordinate systems. Never publish a learner’s work or identity without the relevant permission.

Access options: use on-screen movement controls, adjust speed, provide a partner driver/navigator role and allow spoken reflection. Keyboard play is supported. The canvas world does not yet offer a nonvisual equivalent, so the prototype is not fully accessible.

Source review record and limits

The supplied coordinator evidence identifies partner course 195, Custom Roblox Worlds, and course 201, Adventures in Roblox Studios. An October 19 partner-channel message indicated the latter videos were still being recorded. This is contextual evidence, not a review of either entire course.

On October 30, this task located a credential message in the authorized private Slack DM. No credentials were saved to source or deliverables. The coding platform’s sign-in page was verified. The Slack browser session was signed out; a full authenticated curriculum review was not completed. No lesson sequence, video coverage or quality comparison is claimed from inaccessible material. The public sales page was also unavailable through the web fetch tool.

Before calling this an improved replacement, review both full outlines and available lessons through an authenticated session, record concept coverage and missing content, compare those goals against these original units and revise the transfer tasks. Do not copy partner prose, video transcripts or assets.

References: partner platform; sales page supplied by coordinator; partner setup message.