Hey @elonmusk, sending humans to Mars with AI and mechanical eng. is impressive.
Can we also use AI to create free food and a personal robot therapist for everyone?
We’ll need both on Earth, and even more on Mars.
Do that, and you might become the coolest man in history! ;)
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.
@karpathy People have been saying that knowledge bases and data are the keys to GenAI success, but few have had a practical idea of how to make it happen. But ideas like these can give everyone the missing pieces of the puzzle!
So I've spoken to many Iranians who want intervention to topple the regime and to Westerners who say they absolutely don't
The Iranians win the argument
⚠️⚠️⚠️سازمان ملل متحد با سپردن کرسی نایبرئیسی «کمیته منشور ملل متحد» به جمهوری اسلامی، عملاً "حقوق بشر" را ترور کرد!
✊️چطور میتوان نظارت بر اجرای منشور سازمان ملل را به رژیمی سپرد که خود بزرگترین ناقض بندبندِ آن است؟ رژیمی که نامش با تروریسم، سرکوب داخلی و نقض سیستماتیک حقوق بشر گره خورده، اکنون قرار است برای «تقویت نقش سازمان ملل» تصمیمگیری کند!؟
این انتصاب، توهین مستقیم به خون هزاران قربانی استبداد و تروریسم در ایران است که هنوز آثار قتلعام آن بر کف خیابانهای ایران خشک نشده!
سازمان ملل همدست قاتلین ملل است!
@UN_Spokesperson@UNWatch@UN@volker_turk@antonioguterres@eleanorsanders@drmaisato
من بهتون قول شرف میدم
ارتش آمریکا در 8 ساعت 40 هزار ایرانی رو سلاخی نخواهد کرد
ارتش آمریکا بازار رشت و مردم داخلش رو زنده زنده آتش نخواهد زد
ارتش آمریکا حکم اعدام گروهی به بچه های بیگناه مردم نخواهد داد
ارتش آمریکا جنازه ها رو در سطل آشغال و بیابان ها نخواهد ریخت
#جاویدشاه
Nastaran Panahi was just 16 years old when she was shot in Nasimshahr, Tehran. The same Islamic Republic that killed her is now appointed by the United Nations to a human-rights advisory position on the UN Human Rights Council. Yes, you heard me correct.
This is not “ diplomacy.” This is a regime shooting children.
The people of Iran are no longer just mourning, they are demanding and END for Islamic Republic. They are standing up so that no other 16-year-old girl’s life ends with a bullet.
Witnesses say she was rushed to a nearby clinic, but she began bleeding from the mouth and did not survive.
#IranMassacre
خاتمی با ایستادن در کنار خامنهای و رژیم خونخوارش، تهماندهی اعتبار و احترام خود را از دست داد.
از این پس، اردکان را با جاویدنامان آزاده و شجاعی چون «مصطفی سرافراز اردکانی»، دانشجوی مهندسی مکانیک دانشگاه یزد و جانفدای میهن، خواهیم شناخت.