DesmosGPT

What the company appears to be: DesmosGPT is an AI-powered SAT Math tutor that teaches students how to solve problems using the fastest Desmos strategies. The author states it is built as a prototype and submitted to the OpenAI 2026 hackathon.

What changed: The project evolved from a ChatGPT + Lovable prototype to a more modular system using Codex, GPT-5.6, and AI coding agents (Cursor, Windsurf, Antigravity), with a redesigned UI and scalable technique engine.

The single most important open question: Is there evidence of any traction, revenue or customer adoption beyond the author's own development efforts?

The description states that DesmosGPT is an AI-powered SAT Math tutor focused on speed through Desmos. It allows students to input questions via text or screenshot and generates interactive step-by-step walkthroughs within a live Desmos calculator.

Evidence:

• "Students can type a question or simply paste a screenshot."
• "The AI identifies the problem type, selects the best Desmos technique from its knowledge base, and generates an interactive, step-by-step walkthrough inside a live Desmos calculator."

Inference: The product appears to be a web-based application integrating AI with the Desmos graphing calculator platform.

The author positions DesmosGPT as the first AI tutor that teaches students the fastest Desmos strategy for every SAT Math problem. It claims to focus on speed and workflow optimization rather than generic math explanations.

Evidence:

• "The first AI tutor that teaches students the fastest Desmos strategy for every SAT Math problem."
• "I wanted to build an AI coach that doesn't just solve SAT Math—it teaches students how to think in Desmos."

Inference: The positioning evolved from a general AI math tutor to one specifically focused on optimizing SAT performance through Desmos.

The target customer is SAT students preparing for the Digital SAT, particularly those who struggle with time management during the test.

Evidence:

• "As an SAT student, I noticed that many classmates—including myself—constantly ask our teacher the same question: 'What's the fastest way to solve this in Desmos?'"
• "The next steps are expanding the knowledge base to cover every major SAT Math topic"

Inference: The primary user is a high school student preparing for the SAT, with potential expansion to broader educational markets.

No evidence of business model or pricing structure is provided in the description.

Evidence:

• Not evidenced.

Inference: Based on the author's own account, this appears to be an MVP built during a hackathon without any commercial implementation.

The project was built using React, TypeScript, and Vite with a modular architecture designed for scalability. It integrates AI models (ChatGPT, Codex, GPT-5.6) and supports multimodal input.

Evidence:

• "Built with (author-declared): css3, html5, json, react, tsx, typescript, vite"
• "The result is a modular system where new Desmos techniques can be added as reusable knowledge files instead of hard-coded logic."
• "Implemented multimodal input, allowing students to paste screenshots or type mathematical expressions naturally."

Inference: The technical stack suggests a modern web application with AI integration and extensible architecture.

No evidence of traction, revenue, customers, or adoption is provided beyond the author's own development efforts.

Evidence:

• "This project is still an MVP."
• "The next steps are expanding the knowledge base to cover every major SAT Math topic"
• "I iterated on the product every day."

Inference: The project remains in early-stage development and has not reached any measurable user base or commercial deployment.

No evidence of competitors or market positioning is provided.

Evidence:

• Not evidenced.

Inference: Without additional context, it's unclear whether there are existing tools addressing similar needs in SAT prep or educational AI tutoring.

Key risks include lack of commercial traction, unproven scalability, and dependency on AI model performance. The project is described as an MVP with no revenue or customer data.

Evidence:

• "This project is still an MVP."
• "Even strong LLMs don't always choose the fastest SAT Desmos strategy."
• "No revenue, customer or traction data is available beyond what they state."

Inference: The lack of commercial validation raises concerns about market demand and product-market fit.

1. What specific metrics or outcomes have you observed from using the tool during development?
2. How do you plan to validate that your approach actually improves SAT performance?
3. Have you tested the tool with real students, and what feedback did they provide?
4. What is your roadmap for transitioning from MVP to a scalable product?
5. Are there any partnerships or institutional relationships in place to support distribution?