Alan Chan

The most underrated use of AI isn’t learning faster—it’s becoming capable of learning knowledge that’s deeper, harder, and more abstract. In this article, I share how I use AI to teach myself a textbook step by step, and why the gains in depth + quality have been mind-blowing.

@HanB411971 @Heptabase If you have the PDF file, then yes. We haven't supported EPUB yet, but we plan to support it soon.

@alanchan_tw @Heptabase can this AI tutor also be used in book digestion, e.g. reading a book with this AI tutor

I started @Heptabase 4 years ago, with the mission to give people the power to gain a deep understanding of anything. We began by building a visual knowledge base for students, researchers, and lifelong learners. It quickly became a beloved product used in over 150+ countries around the world, won the @ProductHunt annual Golden Kitty Award in the Personal Productivity category, and scaled profitably into millions in revenue. However, I’m not satisfied. For sure, it’s a great app to organize and make sense of all your learning and research notes. But it’s still far from our mission. You can’t learn quantum mechanics or the theory of relativity just by downloading Heptabase. You still need to have a teacher or spend a lot of time reading difficult sources on your own before you can get the most value out of Heptabase. Things have changed today. I’m very excited to introduce Heptabase’s AI Tutor — an AI agent that is designed to help you achieve any learning goal you’ve set. This is not another “upload and ask” AI app like ChatGPT, Gemini, or NotebookLM. It’s built to help you make step-by-step progress toward your goal: 1. Goal-Oriented Curriculum: Simply state your learning goal (e.g., study Western philosophy, understand international politics, pass the bar exam, or achieve financial freedom), and the AI Tutor will design a personalized syllabus tailored to your level and needs. 2. Systematic & Personalized Learning: Instead of trying to figure out what question to ask, the AI Tutor guides you. It provides step-by-step reading materials—allowing you to highlight important content and ask questions about anything you don't understand—then tests your understanding and dynamically adapts to your learning style and pace over time. 3. Seamless Note-taking Integration: Your highlights, notes, and Q&A sessions are automatically organized onto dedicated whiteboards and converted into clean review notes. You have the best note-taking experience in the world available to you while learning with AI! AI Tutor is another significant product milestone that helps us move toward our mission. In the past two months, I have taken over 50 lessons using AI Tutor, and the experience is just mind-blowing. It’s a dream come true for all curious people who want to understand more about the world. I really wish I had it when I was a kid! AI Tutor is now available for all Premium Plan users and will be available for Pro users soon. Try it out, tell your friends, and share your feedback with us! I hope you enjoy it!

@Heptabase Youtube Link: youtube.com/watch?v=GFiM6h…

不確定你是否參考到了舊版資訊，這邊針對 Heptabase 最新版的功能現況稍作補充。 關於 AI Credit：自 12/5 起，所有 Heptabase 用戶皆可直接使用官方提供的 Monthly AI credit，無需再自備 API Key。 關於 AI 功能：Heptabase AI 的核心優勢在於對 PDF 文獻的深度處理。我們允許用戶上傳任何大小與頁數的 PDF，並能解析完整內容。這帶來了幾個關鍵差異： - 精確引註：AI 回覆時會附上 Citation，點擊即可一鍵跳轉回 PDF 的特定段落與頁數。 - 彈性提問範圍：你不一定要針對整份文件提問，可以在閱讀過程中指定「特定頁數範圍」或「特定段落」與 AI 討論。 - Max Mode 全文閱讀： 即便面對數百頁的文獻，開啟 Max Mode 後，AI 也能讀取「完整、無遺漏」的全文內容。 相較之下，Notion 的 PDF 限制在 20MB 與 100 頁以內，且只能針對整份檔案提問，AI 的回答無法引用回 PDF 的內容；NotebookLM 則缺乏整合的 PDF 閱讀器，無法邊讀邊針對指定內容提問，且模型僅支援最基礎的 Gemini Flash。 可能是基於成本考量，目前不論是 Notion、NotebookLM、Gemini、ChatGPT，都未提供類似 Max Mode 的機制來讓 AI 真正讀完整份長文獻，這在需要深度研讀的場景中是巨大的短板，而這種在「深度學習和研究」的場景中才會遇到的痛點，正是 Heptabase 專注在解決的問題。

最近在重新整理和设计「重新思考」会员知识库，还有重新梳理一些论文和书籍章节，我又回到了 Heptabase 中。 白板在处理这种任务时，异常好用。拆解 PDF、记笔记，再通过连接，作为线索的指引，减轻大脑负担。 如果你不知道如何使用 Heptabase，可以阅读这篇文章：noesisapient.com/p/heptabase。 在这篇文章中，我梳理了人类大脑的分类方式，以及如何结合 Heptabase 中的元素，更好地帮助我们学习。后续，会在栏目内分享详细分享更多内容。 如果你厌倦了传统的文档式笔记软件，不妨试试 Heptabase：join.heptabase.com/?invite-acc-id… Heptabase 的新 Slogan: 「是一款专为学生、研究者和终身学习者打造的智能可视化知识库」 讽刺的是，对于学生和学者而言，有免费的 Notion、Gemini、notebooklm 可以使用，功能要强大的多。所以 AI 在 Heptabase 中显得微不足道了。需要设置 API，使得使用 AI 变得更加繁琐。我更期望 Heptabase 可以注重细节和 UX 的优化，而不是天天在群里搞投票了。 白板应用的 AI 应该可以让用户直接和白板所有物进行互动和再编辑，例如排版、连线、生成新卡片……而不是回到侧边栏，再输入、生成、再拖拽。这种体验是割裂的。

I don’t see the two as mutually exclusive. In this piece I’m just sharing how I bring AI into my reading and note-taking flow. But if I hit something that really needs a hands-on step—like actually working out a problem or running some code—I’ll go do that first, then come back and write what I learned into my notes. For example, when I was reading this textbook there was a part where I wasn’t sure why a certain assumption was made. So I opened up Desmos, plotted it out, thought it through while looking at the graph, and once it clicked I grabbed a screenshot, dropped it into my notes, and wrote down my thoughts.

Hey Alan, this is really great. How do you add in the hands-on part of learning into this workflow? For the pattern recognition example, having some code in my ide to test out concepts would be huge. Copying/pasting my code snippets and connecting them to the textbook would close a big gap. A simpler example is if I’m learning new math concepts. Working through examples is useful, and being able to snap a picture from my phone and sending it to the heptabase app would also be a really great feature.

The most underrated use of AI isn’t learning faster—it’s becoming capable of learning knowledge that’s deeper, harder, and more abstract. In this article, I share how I use AI to teach myself a textbook step by step, and why the gains in depth + quality have been mind-blowing.

@berliangor Yes! I wrote about him in another article: medium.com/heptabase/my-v…

@alanchan_tw This is great! Have you heard of Xanadu Spaces by Ted Nelson? The same idea but was way ahead of his time

@JeremyNguyenPhD I'm using @Heptabase to set up the whiteboard, PDF, and AI chat workflow. It's a software, not a physical electronic whiteboard. There are some community-shared whiteboards in the public gallery if you want to learn more about it. heptabase.com/gallery

@alanchan_tw Is your whiteboard an actual physical electronic whiteboard? I haven't heard many people mention page breaks etc as a form of friction. If be very keen to know more about your set up.

Article Link: medium.com/heptabase/the-…

One of my favorite features!

In the latest version, you can add whiteboards as contexts to AI chat! In AI's response, we’ve introduced reference links to the cards and blocks that the response is grounded on, so you’re able to locate the source content.

Check this out!

@j_sevillejo @Heptabase A much advanced version will come this month.

@Heptabase Is this coming back anytime? I'm really tempted to get Kortex AI by Dan Koe, but I don't want to buy another note taking app. Just really waiting to take two cards, and help me make sense of another card using it. It will be so useful, if you have a guide to follow for example.

In the latest version, we’ve launched Assistant — an AI chatbot that can provide you with the best card recommendations based on your questions, regardless of the language and phrases you use. Please note that Hepta AI does not use your data to train AI models.

This is finally coming!!

@sherlockieee @RichardMCNgo They can definitely be described in @Heptabase.

@RichardMCNgo Which is a shame bc most of the things I want to see in the world cannot be described in a tweet

Tweet the training data you want to see in the world.

★ Dynamicland's new website documents ten years of progress toward a humane dynamic medium. dynamicland.org

Heptabase's daily use case. I always start my day by deciding what I want to work on and writing down my to-dos in the Journal. For example, today I’m focusing on researching “application layer design.” My to-dos include the things I want to read, the questions I want to answer, etc. I’ll then go to the “Product” tab group and enter the pinned whiteboard called “Application Layer” to stare at it and think, add new content to existing cards, add new cards, add new connections, rearrange cards, break down cards, etc. During the process, I might clip some internet content or add some PDFs I found to the whiteboard, break it down into atomic cards, delete useless cards, rewrite the headers of useful cards in my own words, etc. That’s pretty much it. Over the course of a few weeks, the whiteboard will go from empty to messy to structured, and then eventually I might write a document that synthesizes my research on this whiteboard and share this document with my team. When drafting this document, I always open it on the right sidebar of the whiteboard and start by adding relevant links from the whiteboard to the document (using the copy link feature), so I can click those links to navigate the whiteboard and replace those links with the words I write.

Heptabase's plans for whiteboard collaboration. The reason we’re building the Whiteboard: Collaboration and Discussion is the same as the reason we build other feature: we want to better serve our users who’s working on a complex project, which a lot of times involves a lot of discussions and feedback loops that result in a high volume of text responses that are hard to trace back and make use of in the future. We’re less interested in solving the problem of “presentation” and more interested in helping our users gradually build “structure” on top of all the discussions they have around a topic. There will be two phases of implementation. The first phase is rather trivial, which is laying out the technical foundation of the permission model where you can invite other people to your whiteboard, collaboratively edit this whiteboard, let other people bring their own cards to your whiteboard, etc. The second phase is more important, which involves introducing the discussion feature that has a lot of interplay with the whiteboard.

I hosted an AMA in @Heptabase 's Discord community last weekend. Check out this thread to see my responses to the top upvoted questions, which cover the following topics: - Heptabase's position in the PKM market and its attitude toward AI. - Heptabase's attitude toward interoperability. - Heptabase's plans for team expansion. - Heptabase's plans for upcoming feature direction. - Heptabase's plans for diagramming. - Heptabase's plans for long-form writing. - Heptabase's plans for whiteboard collaboration. - Heptabase's daily use case.