Noggimigo
Noggimigo is an AI-powered tutor and patient learning coach built natively into the Noggin platform. Instead of simply providing answers to students, Noggimigo acts as a dynamic instructional guideβscaffolding complex problems, breaking down multi-subject content, and adapting explicitly to students with diverse learning styles and unique needs.
π Key Features
- Coach-Centric Guidance β Guides students toward answers dynamically through hinting and conceptual breakdowns instead of direct solutions.
- Adaptive EdTech Scaffolding β Aligns instructional support with individual progress to support students with learning disabilities.
- Multi-Subject Mastery β Native conversational loops tuned for Mathematics, Science, and Language Arts.
- Contextual Memory β Remembers a student's stumbling points across a session to adjust immediate learning paths.
π οΈ Setup & Integration
Prerequisites
- Python 3.11+
- Dependencies from the core Noggin repository
Quick Start
To initialize the Noggimigo agent workspace within your project environment:
# Clone and enter the directory
git clone https://github.com/Noggin-Labs/Noggin.git
cd Noggin
# Create virtual environment
python3 -m venv .venv
source .venv/bin/activate
# Install the package and dependencies
pip install -e .
π Basic Usage
Initialize the AI Coach
Run Noggimigo directly to start an interactive student simulation or tutor session:
python -m noggimigo.coach --subject mathematics --level entry
Run Student Simulator Validations
To benchmark or evaluate how Noggimigo interacts against student archetypes:
python -m noggimigo.simulator --config profiles/default_student.json
π€ Contributing
We welcome contributions focused on improving the accessibility and educational efficacy of Noggimigo!
- Fork the repo.
- Create your branch (
git checkout -b feature/noggimigo-improvement). - Commit your changes (
git commit -m 'Enhance scaffolding logic'). - Push your branch (
git push origin feature/noggimigo-improvement). - Open a Pull Request.
Inference Providers NEW
This model isn't deployed by any Inference Provider. π Ask for provider support