Mastery

The description states that Mastery is a question-first learning environment. It allows users to import text, select confusing parts, and ask questions. The system responds with explanations and builds a concept map. Concepts are verified through transfer checks and teach-backs, where an AI student (Milo) evaluates the user’s ability to explain the concept in their own words.

The process includes:

• Streaming tutor explanations
• Mapping concepts into nodes
• Generating transfer checks
• Playing Milo, an AI student who grades teach-backs
• Identifying gaps via dashed ghost nodes on the map

It also supports exporting concepts to Obsidian with full evidence history and wiki-links that mirror the concept map.

Inference The product is built around a learning loop involving questioning, mapping, testing understanding, and retention. It is not a general-purpose note-taking tool but a structured study environment.

The description states that Mastery aims to separate recognizing a concept from being able to explain its mechanism, and does so while the text remains open in front of the user.

It evolved from an initial idea of a branching reader into a system focused on transfer checks and teach-backs. This pivot was driven by time constraints but also reflects a stated conviction that studying without checking oneself leaves little behind.

The authors claim that:

• Answering questions helps, but formulating ideas in your own words is what makes them stick.
• The teach-back is the heart of the product, not a bonus feature.
• The evaluator refuses to accept fluent paraphrasing — it requires mechanism-level evidence.

Inference The positioning centers on active learning and retention through structured feedback, rather than passive consumption or note-taking.

The description does not explicitly state who the target customer is. However, based on the use case:

• Users import texts (e.g., books, articles)
• They are interested in deep understanding, not just skimming
• The app supports exporting to Obsidian, suggesting a user base that values structured knowledge management

Inference The ICP likely includes students, researchers, or professionals who want to deeply understand and retain information from texts. It is not described as targeting casual readers or general note-takers.

There is no evidence in the description of a business model or pricing strategy. The project is presented as a hackathon submission with no mention of monetization, subscriptions, or sales.

Inference No commercial structure is evident from the self-reported description.

The product is built using:

• OpenAI GPT models (specifically gpt-5.6-luna)
• Codex for implementation and testing
• React, TypeScript, Vite, PostgreSQL, Supabase
• The system streams responses, maps concepts, generates checks, and evaluates teach-backs

Key technical elements:

• All structured output is validated server-side
• Codex verified GPT guidance, scaffolded the app, and implemented core flows (state model, streaming route, PDF import/export)
• Tests are run end-to-end after every change

Inference The system is built with a focus on AI integration, structured outputs, and automated testing. It appears to be a prototype or MVP, not a production-ready product.

There is no evidence of traction, revenue, customers, or adoption beyond the self-reported project description. The authors note that it was built during a hackathon and that they had limited time, with sessions ending in working prototypes.

The project has:

• A team of two members
• No mention of users, usage metrics, or product-market fit

Inference There is no evidence of traction or maturity beyond the initial prototype stage.

The description does not provide any information about competitors. It does not reference existing tools in the learning or knowledge management space, nor does it explain how Mastery compares to them.

Inference No competitive positioning or market analysis is evident from the self-reported description.

• No traction or revenue evidence: The project is described as a hackathon submission with no commercial data.
• Unverified claims: All descriptions are self-reported and unverified.
• Limited team size: Only two members, which may limit execution speed or depth.
• No pricing or monetization model: No indication of how the product would be sold or funded.
• Unclear scalability: The use of a single AI model (gpt-5.6-luna) and Codex for implementation suggests a prototype-level system.

1. What is your plan to validate the learning effectiveness of the teach-back mechanism?
2. How do you intend to scale beyond the current prototype, especially with AI model limitations?
3. Are there any early users or test subjects who have tried the system?
4. What is the long-term vision for monetization or product-market fit?
5. How does Mastery differentiate from existing tools like Anki, Obsidian, or Notion in terms of learning outcomes?