aita-core

Shared core package for AI Teaching Assistant (AITA) chatbots.

aita-core provides a complete, course-agnostic framework for building AI teaching assistant chatbots. It includes a Streamlit chat UI, RAG (Retrieval-Augmented Generation) pipeline, document ingestion, interaction logging, and an admin dashboard — all parameterized by a single CourseConfig object.

Key design principle: the chatbot never gives direct answers. It guides students through Socratic questioning, hints, and conceptual explanations.

from config import CONFIG
from aita_core import run

run(CONFIG)

Features

  • Pedagogical guardrails — Never gives direct answers; uses Socratic questioning

  • Week-aware responses — Won’t discuss topics not yet covered in class

  • RAG pipeline — Retrieves relevant course materials to ground responses

  • Source citations — References specific handouts, slides, and homework with PDF downloads

  • Admin dashboard — View interaction logs, student feedback, and manage course settings

  • Google OAuth — Restrict access to @umn.edu accounts (optional)

  • Wikibook ingestion — Automatically fetch and index online textbook chapters

  • Mobile-friendly — Responsive UI works on phones and tablets

User Guide

Architecture

API Reference