Wei Huang(mrhuangtalk.com)

前程序员，现在是IT产品经理和努力为财务独立搬砖的个人投资者。

我的个人网站 Mr.Huang Talk ( 主要分享我在如何提升“工作收入”和“被动收入”方面的思考和感悟

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@yibie 可惜搜索没用llm

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@antoniayly 正好需要

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

https://t.co/e9NICZ7b84 我把cladue code源码让ai写了本10万字的书，还建了个llm wiki，可以问答

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

真是太有意思了，最近被这个凸性资产上头，我就用芒格去挑战塔勒布，这句话让我很有感觉

塔勒布想让自己在任何未来中都有选择。我想让自己在任何未来中都不需要选择。

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

一个小时把超越百岁所有章节编译成 wiki，效果还不错。

[Andrej Karpathy @karpathy]

LLM Knowledge Bases

Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So:

Data ingest:
I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them.

IDE:
I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides).

Q&A:
Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale.

Output:
Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base.

Linting:
I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into.

Extra tools:
I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries.

Further explorations:
As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows.

TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@qkl2058 这是 3 台 mac mini ？🤣

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@berryxia 现在pdf转markdown最好的是哪个？

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@yibie 可惜搜索没用llm

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

llm wiki

[yibie @yibie]

大力推荐这个网站！它把巴菲特致股东的信掰开了，揉碎了，结构化了，变成一个知识库。而我非常喜欢它的结构：总览，图谱，概念，索引。

巴菲特致股东信知识库
https://t.co/Rkt9gy08dj

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

我用小龙虾代替 Obsidian webcliper，效果很好。我的小龙虾用 inbox 存放 raw，再用 LLM wiki 管理

[Andrej Karpathy @karpathy]

LLM Knowledge Bases

Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So:

Data ingest:
I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them.

IDE:
I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides).

Q&A:
Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale.

Output:
Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base.

Linting:
I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into.

Extra tools:
I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries.

Further explorations:
As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows.

TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@imedwardluo 老板审美在线，很有设计感

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@gkxspace workbuddy可以操作微信文档

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

豆包的 coding plan 用的 kimi 2.5

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

早知道阿里coding plan那么卡，买Lite就行了，还是用的qwen3.5plus，要是用kimi、minimax不是更卡了

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@wa75303363 才用的时候挺好的，今天下午卡不行，我换成kimi试试？

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

经过8843mins，pre train完成

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

用我16G 的 4070tis 训练 nanochat，感觉要跑好几天

#nanochat

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

4070 3.5 天完成了 8 卡 H100 2 小时的任务

[Wei Huang(mrhuangtalk.com) @MrHuangTalk]

@GitHub_Daily 这个和扣子比，有什么优点？