BuckeyeQuest

BuckeyeQuest helps elementary teachers plan a grade-aware learning day, gives students teacher-assigned interactive missions, and turns their choices into grounded support recommendations.

Comment

BuckeyeQuest

Category

Education

What it does

BuckeyeQuest is an AI-powered lesson planning and student engagement platform for elementary teachers in Ohio. It solves two interconnected problems:

For teachers: Turning a grade-level learning goal into a structured, actionable day plan to help make sense of student choices to inform next steps.

For students: Delivering grade-appropriate, teacher-assigned learning missions across reading, math, science, history, and creative subjects, with an interactive narrator that reacts to their choices in real time.

Teacher workflow:

  1. Input grade level, Ohio learning priorities, subjects, dates, and a specific learning goal
  2. Receive a reviewable full-term plan with Monday through Friday rhythm and lesson-level details
  3. Adjust the plan, choose guided or free-choice journeys, control hint visibility, and assign each learner's grade
  4. Monitor student progress through the dashboard
  5. Ask Buckeye Assistant: ask any quesitons about the class and get evidence-based recommendations

Student experience:

  • Join via class QR link
  • See only their teacher-assigned, grade-level missions
  • Choose from subject-based missions (not a generic game list)
  • Interact with Buckeye Narrator, which reacts to each choice
  • Write alternate paths or ask a journey tutor for help

Key differentiator: Teachers remain in control. BuckeyeQuest can suggest challenge-level changes after strong performance, but the teacher must review and approve every adjustment.

How we built it

Architecture:

  • Frontend: React PWA with local-first architecture
  • Backend: Node.js with Express
  • AI Runtime: NVIDIA Llama 3.3 Nemotron Super 49B v1.5 for live inference
  • Development AI: GPT-5.6 and Codex for rapid prototyping

Key technical decisions:

  • Local-first design: The app remains playable when AI providers or networks are unavailable
  • Security-first: Browser never receives an API key; student names anonymized before teacher-assistant requests
  • Deterministic fallbacks: Preserve the demo loop even without connectivity
  • Teacher approval gate: Differentiation changes require manual review before taking effect

Challenges we ran into

  • Balancing AI-generated personalization with teacher control: solved by making all differentiation suggestions reviewable and approval-gated
  • Maintaining offline functionality while using cloud AI: solved with deterministic fallback narratives and cached responses
  • Anonymizing student data effectively for AI queries while preserving enough context for useful recommendations
  • Designing a branching narrative system that feels responsive and educational, not generic
  • Ensuring grade-appropriate content across five subjects without hardcoding every possible path

Accomplishments that we're proud of

  • Built a working teacher-student loop where choices actually inform next-day planning
  • Created an offline-capable PWA that doesn't break when AI is unavailable
  • Designed a QR-based classroom handoff that takes seconds to set up
  • Made teacher review a feature, not an afterthought, and keep the educators in the driver's seat

What's next for BuckeyeQuest

  • Expanding subject mission libraries with more Ohio-specific content and maybe across different states
  • Adding parent-facing progress summaries
  • Building a collaborative planning feature where grade-level teams share and adapt plans
  • Implementing voice-based interaction for younger students
  • Developing analytics dashboards that surface class-wide trends and intervention opportunities
  • Exploring integration with existing LMS platforms like Canvas and Google Classroom

Built With

Share this project:

Updates