Charles Ai

@elonmusk U guys using ratatui for rending im guessing and that must’ve been annoying lol

Grok Build

@elonmusk lol at least u guys got the rust part right

@elonmusk lol oh boy

Try this early Grok Build (anything) beta and let us know what to improve. Much appreciated!

@obsdmd Sure I’m down lol

The Obsidian team is growing from three engineers to four engineers. Competitive SF salary. Fully remote, live anywhere. Apply below.

@jasonkneen I mean the guy has Mac’s days more than enough lol

Good god please don't listen to this horseshit. 1. OpenClaw is a ralph loop with messaging it's not f**king the greatest piece of software ever written and there's alternatives like paperclip and nanoclaw and tinyclaw that work as good if not better. 2. You don't have to be burning money like this fool. He's making his on clicks and views and so all this drama is great for his numbers but having a 24/7 workforce that's cost him tens of thousands has lead to him releasting ZERO products over last months. Even his screenshots showed 0 tool use. It's all engagement farming bullshit. But you can choose to follow the fool or listen to people who actually know what they're talking about and aren't farming for likes and replies and horseshit. Up to you I have no skin in the game -- just sick of FUD.

@johnennis Same ur welcome to talk to me anytime

I think one of the biggest challenges when it comes to going hard into using AI is loneliness I am learning all these awesome things and becoming super capable But the set of people that I can really talk to about it is very small Is anyone else having this experience?

lol I’m building a rag anything system the concept is actually very similar to yours but instead we are using postgre to ingest all the data. It’s interesting what you have here because this essentially gives me the lenses to peep into the database without queries. There are obviously downsides to this as well as once it goes over a certain amount of mds u might have issues. Database is still the way to go with giving this thing what it needs but love the concept only thing I would say is a daemon or cron that does it automatically .

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.

@AlexFinn Ur so dramatic

It’s over. Anthropic just banned OpenClaw. Uncensored thoughts: 1. Massive mistake that will come back to bite them 2. Open source needs to win. If you have a local model running on your Mac mini, no corporation will ever be able to ban you 3. ChatGPT 5.4 is the best model. But it sucks compared to opus in OpenClaw. I will continue to pay for Anthropic api 4. I have no doubt the next OpenAI model will be optimized for Openclaw and be excellent 5. In 6 months the local models will be as good as opus 4.6 and all of this will be forgotten 6. It’s feels like from a consumer sentiment perspective things have flipped for OpenAI and Anthropic. They were the darlings when Opus 4.5 came out 7. Going to the Kanye concert right now please don’t spoil the stage or set list in the replies 8. The best openclaw set up is now Opus as the orchestrator, then much cheaper models as the execution layer. If you do this properly you won’t be paying much more than $200 a month. I’m using Gemma 4 and Qwen 3.5 for execution on my DGX Spark and Mac Studio

@NetworkChuck Local is the way trust no one

Anthropic says NO MORE OpenClaw, they are officially cutting us off starting April 4th. How do you feel about this? Will this hurt Anthropic, OpenClaw or both?

@elonmusk Give us a real harness and I’m ready to jump ships

Current release of Grok is much than beta 1, which beat Opus in this arena. We usually update the model twice a week. Try the current version of Grok Heavy and you will be pleasantly surprised.

@automate_archit @bcherny Sure if it’s actually one shots everything

@bcherny Honestly this pushes people toward the API faster — which is where you get real control anyway. Most devs I work with were already hitting subscription limits on heavy Claude Code sessions. API keys + usage bundles = pay for what you actually use. Better economics at scale.

Starting tomorrow at 12pm PT, Claude subscriptions will no longer cover usage on third-party tools like OpenClaw. You can still use these tools with your Claude login via extra usage bundles (now available at a discount), or with a Claude API key.

@bcherny Fix cowork before you even attempt this . Cowork is clunky , times out and just shitty period

@bcherny @jaredctate U mean copy them

We're big fans of open source. I actually just put up a few PRs to improve prompt cache efficiency for OpenClaw specifically. This is more about engineering constraints. Our systems are highly optimized for one kind of workload, and to serve as many people as possible with the most intelligent models, we are continuing to optimize that. When you use an API key or overages it should still work. The issue was just subs. If you still want to cancel, we're giving full refunds. We know not everyone realized this isn't something we support, and this is an attempt to make it clear and explicit.

@bcherny I guess we going to openai then.

@krispuckett @openclaw @claudeai Incredibly fast

Anyone else using @openclaw with Codex? How is it compared to @claudeai ? Seriously considering canceling Max plan now.

Sigh anthropic ur forcing me to break up with you. I guess hello to openai lol

@realsigridjin lol i see your having real issues with ratatui not suprised.

@LexnLin unforunately at the end of the day even if you clone this thing your constraint by context

I gathered ALL claude code's system prompts/prompts for agents in a repository: github.com/Leonxlnx/claud…

@rohan_2502 @AnthropicAI @claudeai lol closed source doesnt mean you have access to their api side.. if you built cli you will understand there is alot of missing parts that this thing doesnt show yet

don't worry @AnthropicAI , am handling the @claudeai code fixes for you. will push the fixed code on github, can fork if you think the changes makes sense. let's see how it goes, excited.

@Austen are u going to sell my kidneys?

Engineers! We want to: Fly you to Austin Give you an apartment Clean your room Do your laundry Feed you 3x/day Help you master building with AI And give you a $200k to $1m job. You pay nothing, ever, no matter what. Even if you don’t take the job.