Weibei

• The description states that Weibei is a macOS-native application.
• It supports reading PDFs, HTML pages, and Markdown files.
• Users can select text, ask questions (via an embedded "study agent"), and link answers back to the source.
• It integrates writing capabilities using Milkdown for Markdown notes.
• Features include immersive view modes, local search via SQLite and Vision OCR, and a course-folder-based workflow.
• The product is built with SwiftUI, AppKit, PDFKit, WebKit, Milkdown, SQLite, Vision OCR, and an embedded Pi runtime for the study agent.

• The description states that Weibei is designed for “macOS native reading chatting and writing.”
• It positions itself as a tool to help users learn by integrating reading, chatting (via AI), and writing into one workflow.
• The author notes that the original idea came from personal frustration with switching between Chrome and Obisidian during study sessions.
• During Build Week, the product evolved to include:
  • A course relationship workbench
  • Persistent Q&A across file types
  • Source-linked answers
  • Foundation for constrained interactive learning answers

Inference The evolution suggests a shift from a simple tool to one with more AI-integrated features and structured learning workflows.

• The description implies that Weibei is intended for students or learners who read PDFs, HTML pages, and Markdown files while taking notes.
• It targets users who want to reduce context-switching between reading and writing tools.
• The author’s personal experience with cramming for exams and using Obisidian suggests a focus on academic or self-directed learning workflows.

• The description does not mention any pricing model, monetization strategy, or revenue streams.
• There is no indication of whether Weibei will be sold as a paid app, offered via subscription, or distributed for free.
• No evidence of partnerships, licensing, or distribution channels.

Inference The author may be developing the tool primarily for personal use, with potential future monetization plans not yet defined.

• The app is built using SwiftUI and AppKit.
• It uses PDFKit and WebKit for reading, Milkdown for Markdown editing, SQLite and Vision OCR for local search.
• An embedded Pi runtime supports the "study agent."
• The author reports using Codex (98% of development) and other AI models.
• During Build Week, 57 commits were made to an integration branch.

Inference The app is a native macOS application with strong integration between reading, writing, and AI-assisted features. It appears to be in active development.

• The description states that Weibei was submitted to the OpenAI 2026 hackathon.
• It includes a history of development with 57 commits during Build Week.
• The author reports personal use and satisfaction with the tool.
• No evidence of external users, customers, or adoption metrics.

Inference The product is in an early stage of development, likely focused on personal utility and experimentation rather than market traction.

• The description does not reference any direct competitors.
• It implies a niche in study tools that combine reading, writing, and AI assistance.
• It is positioned as a macOS-native alternative to tools like Obisidian, Chrome, and other note-taking or reading apps.

Inference Weibei may compete with tools used for academic workflows, such as Obsidian, Notion, or browser-based reading tools. However, no explicit competitive analysis is provided.

• The product is described as a solo effort (1 person team).
• It is not yet commercially viable or adopted, based on the lack of user data or revenue.
• The author notes that the agent part was hard to design and that Codex sometimes proposed over-designed solutions — suggesting technical challenges in AI integration.
• UI complexity was an issue early on, indicating potential UX issues.
• No evidence of monetization strategy or long-term business model.

Inference The project may be at risk due to limited resources, lack of external validation, and unproven commercial viability.

1. What is the actual user base for Weibei? Is it used by others beyond you?
2. How do you plan to monetize the product? Is there a pricing model in mind?
3. What are the key technical challenges you've faced with the AI agent, and how are you solving them?
4. Do you have any plans for expanding beyond macOS (e.g., Windows, web)?
5. How do you intend to validate or improve the user experience beyond your own use case?