LearnLoop - the AI guru that guides you to learn

The description states that LearnLoop is an AI learning coach made of three agents:

• Planner: decides how to teach each question (hint, question, worked example, or next step)
• Coach: gives responses without caving even if the student pushes for a direct answer
• Assessor: checks understanding and updates progress

The system uses GPT-5.6 to power these agents, with a chat interface built in Streamlit and a database using SQLite. It tracks whether students solve problems independently or need full answers, feeding data into dashboards for both students and instructors.

Evidence The author's own write-up describes the architecture and functionality.

Inference The system appears to be a prototype built during a hackathon, not a production-ready product.

The description states that LearnLoop is positioned as an AI coach that "guides students with hints and questions" rather than just answering directly. It claims to address a 2025 MIT study finding that AI use weakens learning and memory, by using the same AI to coach instead of answer.

Evidence The author's own write-up includes the claim about the MIT study and the shift from blocking AI to coaching with it.

Inference This positioning reflects an attempt to solve a perceived problem in education technology — over-reliance on AI for direct answers — but there is no evidence that this approach has been validated or tested beyond the hackathon context.

The description states that LearnLoop targets students who are learning, with dashboards for both students and instructors. It also mentions that it tracks real independence over time and shows which topics need support.

Evidence The author's own write-up describes the intended users as students and instructors.

Inference There is no evidence of a defined ICP beyond general student/instructor roles; no specific segment or persona has been identified.

Not evidenced. The description does not mention any pricing model, monetization strategy, or business model.

The system was built using:

• Codex (for building the agents)
• GPT-5.6 (to power Planner, Coach, and Assessor)
• Streamlit (for UI)
• SQLite (for database)
• Python (programming language)

It was developed in one continuous Codex session over a hackathon week.

Evidence The author's own write-up details the tech stack and development process.

Inference This is a prototype built quickly, not a scalable or production-ready system. The team used a single developer and limited tools.

Not evidenced. There is no mention of users, customers, revenue, or any form of traction beyond the author's own account.

Not evidenced. No information is provided about competitors or market context.

• Unverified claims: The MIT study cited is not independently verified.
• No traction or users: The project is described as a hackathon prototype with no evidence of adoption or feedback.
• Single developer: Only one team member is mentioned, which raises questions about scalability and depth of development.
• Prototype nature: The system was built in a week-long hackathon and has not been tested beyond that scope.

1. What specific feedback have you received from students or educators using this tool?
2. Have you conducted any user testing beyond the hackathon environment?
3. How do you plan to scale beyond a single developer and a hackathon prototype?
4. Is there any evidence of interest from schools, institutions, or educational platforms?
5. What are your plans for monetization or business model?