Caffeine Matrix (@CaffeineMatrix) header
Caffeine Matrix (@CaffeineMatrix) profile picture

iOS Developer since 2015 🚀 | UI/UX Enthusiast

Joined April 2023
Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)Photo by Caffeine Matrix (@CaffeineMatrix)
Caffeine Matrix retweeted
Wow, this tweet went very viral! I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs. So here's the idea in a gist format: gist.github.com/karpathy/442… You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directions to take this in. And ofc, people can adjust the idea or contribute their own in the Discussion which is cool.
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.
1,121
2,825
685
26,773
7,354,318
Caffeine Matrix retweeted
The most dangerous lie in human history isn’t about food. It isn’t about medicine. It is about sleep. For 200,000 years, humans did not sleep 8 hours. That number was invented in 1938 by a mattress company called Simmons Beautyrest. Before that campaign, the average human slept in two shifts. Historians call it “Biphasic Sleep.” You would sleep for 4 hours, wake up for 2, then sleep for another 4. During that 2-hour window, people would pray, have s*x, write, think, and connect with their families. Some of the greatest works in human history were created in that sacred middle window. Shakespeare wrote most of his plays between 1AM and 3AM during his second wake period. Mozart composed entire symphonies in what he called “The God Hours.” Then the Industrial Revolution needed workers on a fixed schedule. You cannot run a factory on biphasic sleep. So they hired a psychologist named Dr. Nathaniel Kleitman to “prove” that 8 consecutive hours was the biological standard. He faked the studies. He was funded entirely by the mattress industry. And the medical establishment adopted his research without question because it aligned with the factory model. They turned the most creative 2 hours of human consciousness into a “sleep disorder.” They called it “Insomnia.” They medicated it. They gaslight an entire generation that 8 hours of continuous sleep was healthy. They pathologized the exact window of consciousness that produced some of the greatest art, music, and literature in human history. You are not an insomniac. You are experiencing the most natural form of human consciousness. And a mattress company convinced you it was a disease. Stop medicating your genius. Wake up at 2AM. Write the thing. The “God Hours” are calling. ✨🙌🏾💫 © Andre Gonzalves
Readers added context they thought people might want to know
Biphasic sleep,two shifts with a wakeful period was historically common.However, “8 hours was invented in 1938 by Simmons Beautyrest” isn’t supported. The post repeats a viral claim with no historical backing (no 1938 ads or campaigns “inventing” the number appear in records). newyorker.com/magazine/2013/…
1,989
8,804
942
31,143
4,516,668
Caffeine Matrix retweeted
It's the year 2003. Everybody is playing games on tiny screens on their mobile cell phones. It's a big business - developers rushing to create more games for all the new devices. We are four years away from the launch of the iPhone. Mobile game development before the iPhone was a totally different show than it is today. What you know from XCode today is the most luxurious development environment I've ever seen. Using the official device simulators from Nokia, Motorolla, Sharp, etc. (when you get them to work) was nice, but it didn't guarantee that your installer file worked the same on the actual real device. Testing all games on real devices before pushing them to the carrier content manager portals was a must. But testing on real hardware means you have to get the actual hardware in your hands. Nokia was nice enough to send us all new handsets before they were released, but often enough, you couldn't get all devices - especially those only released in certain countries like China. Most devices were different in the case of memory and error handling or didn't run at all in other countries. Installing the files on the device wasn't easy either. You couldn't connect it with a cable or Wifi - only some devices did this - you have to set up a SIM + data plan and download the installer file from your own server. At that time, the files were not digitally signed in any way - there were no security setups. You could install anything! When the game was running as expected, we uploaded the content to portals like Jamba, Vodafone, T-Mobile, and other carriers, where they were tested again before they went live on the customer portals. At Vodafone, only one guy developed the complete CMS to handle all this. Vodafone and other carriers were overrun by the mobile gaming hype and had to build their own content delivery platform literally overnight. Wild times! Image showing simulator shot from the official Nokia SDK 2003, running one of my games on a Chinese simulator setup. #developers
2
2
5
401
Caffeine Matrix retweeted
“Google or Apple can build it” No they can’t They have -decades of tech debt -existing products they can’t change without risking billions in revenue -Massive orgs with bureaucratic bullshit -Employees & CEOs who aren’t personally motivated by the problem & don’t give a fuck
49
98
23
1,147
124,010
Caffeine Matrix retweeted
Entrepreneurship isn't risky. Risky is: - Working a job you hate. - With people you don't like. - On problems you don't care about.
20
18
3
140
10,171
Caffeine Matrix retweeted
A git commit a day keeps a 9 to 5 away
43
37
4
560
37,201
Caffeine Matrix retweeted
The greatest trick that venture funds played on entrepreneurs is to make it seem glamorous to spend your most youthful years having meetings with middle managers while you patiently wait for a liquidity event that will arrive when you’re old, overweight, ugly and in the market for a minivan.
58
111
17
2,115
1,149,839
Caffeine Matrix retweeted
Ladies & gentlemen, I present to you... "The Best Moments of Max Keiser" Whether you love him or hate him, Whether you think he's a lunatic or a genius, The FACT is he's been right about #bitcoin since day 1 and you should've listened to him 10 years ago. Max, you are a true OG.
166
417
56
1,934
312,759
Caffeine Matrix retweeted
Another perk of quitting caffeine is I get an extra 2 hours in the day My time asleep has reduced from 9hrs → 7hrs I think because I get 7 hours of uninterrupted deep sleep now and don't wake up to use the bathroom or think about random stuff at 2am
7
1
10
2,846
Caffeine Matrix retweeted
Running bitcoin
11,707
44,242
14,262
132,887
You're too late for whatever has been done. You're right on time for whatever is coming.
4
1
1
74
8,974
✨ Today marks 10 years to the day that Ross Ulbricht was arrested, taken from home, and locked in a cage A first-time offender, arrested for the crime of creating an online marketplace Free Ross. Free #Bitcoin
141
583
43
2,701
191,423