AI Study Assistant

A Streamlit-based AI study assistant designed to help engineering students understand problems and validate their reasoning using open-source large language models.

The app provides two core capabilities:

1. Step-by-step explanations of engineering problems.
2. Automated reasoning verification that identifies mistakes and explains how to fix them.

---

Features

• Explain Mode:

  • Generates clear, step-by-step explanations.
  • Defines variables, shows units, and highlights common mistakes.
  • Adjustable explanation depth (concise vs. detailed).

• Verify My Reasoning:

  • Evaluates a student’s attempted solution.
  • Flags the first incorrent step.
  • Explains what was done well and what went wrong.
  • Provides a corrected solution outline and sanity checks.

• Subject Selection:

  • Allows students to choose from a variety of engineering topics, including as calculus, physics, statics, dynamics, circuits, materials, and thermo.

• Well-Formatted, Easy-to-Read Responses:

  • No chain-of-thought leakage.
  • Markdown formatting.

---

Tech Stack

• Python
• Streamlit (UI)
• Ollama (local LLM inference)
• Open-Source LLMs (llama3.2)

---

Why Local Inference?

The project initially explored hosted inference options, but ultimately uses local inference via Ollama to:

• Avoid API quotas and usage limits
• Enable unrestricted experimentation
• Keep the project fully free to run

This design also has the ability to easily switch inference backends based on cost and reliability tradeoffs.

---

Getting Started

Prerequisites

• Python 3.10+
• Ollama installed

1. Pull the Model:

ollama pull llama3.2

2. Install Python Dependencies:

pip install streamlit requests

3. Run the App:

streamlit run app.py

The app will open automatically in your browser.