coLearn-AI 2026

Future Directions and System Roadmap

While the Summer 2026 work focuses on building a research-grade platform for studying AI-guided learning and epistemic traces, the coLearn-AI system is designed to evolve into a broader environment for AI-supported education.

This page outlines key ideas and planned directions for future development.

1. Activity Mode (Core Learning Environment)

Activity Mode is the foundation of coLearn-AI, where students engage in structured, collaborative learning tasks.

AI-Assisted Activity Creation

  • Interactive activity creator driven by AI
  • Instructors provide prompts; system generates structured activities
  • Iterative refinement and validation of activity design

Timed Suggestions and Adaptive Guidance

  • Configurable timing system for AI interventions
  • Features include:
    • Time reminders during work
    • Suggestions triggered after periods of inactivity
    • Increasingly specific guidance over time
  • Instructor control panel for:
    • adjusting timing thresholds
    • controlling intervention intensity

Configurable Guidance System (YAML-Based)

Move AI behavior out of code and into structured configuration:

  • Platform-level guidance
    • Default system behavior
    • Editable by root/admin users
  • Course-level guidance
    • Each course has its own guidance.yaml
    • Instructors can customize tone, style, and strategy
  • Activity and question-level guidance
    • Fine-grained control over prompts and feedback

Role-Based Guidance

Provide targeted guidance based on student roles within a group:

  • Facilitator: prompts for coordination and discussion
  • Recorder: prompts for clarity and completeness
  • Other roles as defined by the activity

This guidance can:

  • be emphasized early in the course
  • gradually taper as students internalize roles

Expanded Programming Support

  • Extend beyond Python/C++ to include:
    • JavaScript (initial target)
  • Enable broader applicability across courses

2. Test Mode (Authoring and Evaluation)

A dedicated mode for instructors to design and validate activities.

Features

  • AI-generated question creation from prompts
  • Suggested answers and solution paths
  • Instructor testing interface:
    • simulate student responses
    • evaluate AI feedback behavior

Purpose

  • Improve activity quality before classroom use
  • Support rapid iteration of instructional materials
  • Provide a controlled environment for refining AI guidance

3. Demo Mode (Dynamic Presentation Environment)

A flexible mode for demonstrations and exploratory learning.

Features

  • Instructor defines a demo goal
  • System generates evolving content in real time
  • Supports:
    • live classroom demonstrations
    • conference presentations
    • exploratory examples

Purpose

  • Communicate concepts interactively
  • Showcase AI-guided learning in action
  • Support rapid prototyping of ideas

4. Project Mode (Extended, Open-Ended Work)

Project Mode represents a major future direction: moving from structured activities to more open-ended problem solving.

Vision

Allow students to:

  • create or extend multi-file projects
  • write, edit, and execute code
  • test their work using integrated tools
  • receive AI feedback on design and implementation

Potential Features

  • File system interface (create/edit files)
  • Code execution and testing environment
  • Submission and version tracking
  • AI-assisted code review and feedback
  • Automated or semi-automated grading support

Key Design Tension

A central challenge in Project Mode is balancing:

  • Structure
    • predefined assignments
    • guided progression
    • clear evaluation criteria

with

  • Flexibility
    • student-driven design
    • open-ended exploration
    • authentic project work

Future work will explore models that combine:

  • guided scaffolding
  • flexible project development

Summary

These directions represent the broader vision of coLearn-AI as a platform for:

  • structured learning
  • adaptive AI guidance
  • collaborative problem solving
  • and eventually, open-ended project development

While not all of these features are part of the Summer 2026 implementation, they guide the long-term evolution of the system and inform ongoing research and design decisions.

Scroll to Top