Maxwell is a circuit theory study companion, essentially a “Photomath” for electrical circuits. Powered by NVIDIA’s Nemotron Nano VL Vision-Language Model (VLM), it assists Electrical Engineering students in analyzing hand-drawn circuits and calculating complex values such as equivalent resistance, power consumption, impedance, and more.
This project was developed during NVIDIA’s Agents for Impact Hackathon at Howard University. The goal was to build a tool that helps students better understand circuit theory while making a lasting positive impact on STEM education.
- Frontend: React.js
- Backend: Node.js
- Vision-Language Model: NVIDIA Nemotron Nano VL via NIM
- Image-based circuit analysis
- Step-by-step explanations for calculations such as resistance, power, and AC parameters
- Circuit theory quiz mode for knowledge testing
- Feedback on which topics to review or improve
The long-term goal is to make Maxwell a more comprehensive study assistant, while continuing to leverage models from NVIDIA’s NIM platform.
A mobile version would make it easier to capture circuit images and study on the go.
Adding user accounts will allow personalized analysis and progress tracking to help users identify strengths and weaknesses in specific electrical topics.
Here's the demo video: https://youtu.be/NVqEMrtUMgo