Everett Inventor AcademyMission One Β· 20-minute adaptive prototype
β‘0π Lvl 1
β± 20:00
Inventor Mode Β· V5
Train your brain like a spaceship.
Everett, your mission is to help Bolt calibrate a learning engine. It watches what feels easy, hard, fun, or boring β then reshapes the next challenge around you. The engine needs your feedback as much as your answers.
Designed for one satisfying 20-minute session. Wrong answers are signal, not failure. No streaks, no loot boxes, no pressure to return.
β¨ Daily Spark
Tap to reveal β’ No tracking β’ Pure curiosity
Warm-upπ Review
Mission
Question
π§ Good data! Want to try again with what you know?
Choose an answer β I'll adapt based on what happens.
π€
Hi! I'm Bolt. I'll watch and adapt. Tap an answer, then tell me how it felt.
Learning Engine
Calibrating challenge, speed, and confidence.
0
combo
Journey
Train the tutor
Your taps matter as much as your answers.
Session Complete
Engine Calibrated
0%accuracy
Great first mission.
π
Rocket Ratio Lab
Bolt needs your help mixing rocket fuel. Every engine needs the right ratio of spark powder to moon juice. Too much of one, and the engine sputters. Too little, and it won't ignite. Can you crack the ratios?
π Discovery Journal
Facts you've unlocked across sessions
π
Welcome, Inventor!
Pick a mission to start. Each one is a different kind of puzzle β math, science, patterns, or chemistry. You can switch anytime.
π€π―
Know Your Know
Before each question, rate how sure you are. It's okay to guess! This builds metacognition β thinking about your thinking. It's one of the most powerful learning tools.
π¬π
Your Feedback Shapes It
Tap βFun", βToo hardβ, βToo easyβ, or βBoringβ anytime. The engine adapts in real time. Wrong answers are good data, not mistakes. Let's go!
Parent Dashboard
Visual Progress Report
A visual summary of learning across sessions. All data stored locally in this browser only.
Parent Brief
What this is, and what it isn't
A 20-minute prototype of an adaptive learning experience. Built around motivation science, cognitive science, and healthy design. It watches for ease, difficulty, fun, and boredom β then reshapes the next challenge in real time. The goal is to leave the learner wanting to return, not grinding to stay.
What this prototype does NOT do
No streaks. Missing a day costs nothing. No loot boxes. Nothing is randomized behind paywalls. No shame mechanics. Wrong answers produce information, not judgment. No endless feeds. The session ends with a satisfying "engine calibrated" moment at 20 minutes.
About "Bolt"
Bolt is a lightweight visual companion β a mascot, not a pretend friend. It reacts to performance and feelings. It does not claim emotions, sentience, or personhood.
Key Features
Confidence Prediction: Before answering, learners rate how sure they are. This builds metacognition β thinking about thinking. Research shows predicting your own knowledge strengthens learning.
Progressive Hints: Two optional hints per question. The first is conceptual (a thinking strategy). The second is directional (narrows choices). Hints are never penalized β using them is smart engineering, not cheating.
Retry System: After a wrong answer, learners can try once more with a hint. This turns mistakes into genuine learning moments, not just correction events.
Mission Briefings: Each mission starts with a narrative context β why these skills matter. Story framing increases engagement without manipulation.
Adaptive Pacing: The system tracks thinking time per question and adjusts mix accordingly. Quick correct answers signal ease; long thinking + correct signals appropriate challenge; quick wrong answers signal confusion.
Sound Effects: Optional Web Audio API tones for correct/incorrect/level-up. No audio files needed. Can be muted at any time.
V5 New Features
Cross-Mission Interleaving: The "Mixed Mission" shuffles questions from all four topics together. Research shows interleaving β switching between related skills β produces stronger, more transferable learning than blocked practice.
Onboarding Tutorial: First-time users get a gentle 3-step walkthrough explaining missions, confidence prediction, and the feedback system. No pressure, fully skip-able.
Daily Spark: A rotating curiosity question on the home screen, purely optional, no tracking. Designed to build wonder, not obligation.
Visual Parent Dashboard: A visual summary of progress across sessions: accuracy trends, confidence calibration, thinking patterns, and mission breakdown. More accessible than raw JSON.
Bolt Reactions: The companion now visually reacts to correct (bounce), wrong (droop), and combo (celebrate) answers. Expressive, not performative.
Visual Manipulatives: Ratio and fraction questions show visual proportion bars in the scene, making abstract concepts concrete.
Smooth Transitions: Questions now fade and slide between instead of hard-cutting, reducing visual jarring.
Data & Privacy
All data stays in this browser's local storage. Nothing is transmitted to a server. No analytics, no tracking, no accounts. No network requests of any kind. The "Copy parent report" button generates a JSON snapshot you can review or share. The report includes: accuracy, confidence calibration, thinking time patterns, hint usage, retry usage, and learning trajectory.