Chim

does anyone else feel like claude code is pushing using an anthropic api key for any and everything

pov: life after replacing your entire stack with postgres

@ryandcrypto boomers complaining about crypto rug pulls when they're about to execute the most elaborate rug pull in human history

wait so the boomers want us to buy here?? starting to feel like exit liquidity

The Moby Dick fandom is dying. Retweet if you're a true Dickhead.

@Bencera the idea of the "employee" is just gonna go away

The next $1 trillion startup will sell AI employees. Not tools. Not copilots. Employees

@businessbarista @da_fant you nailed it. the biggest challenge is not drowning in slop. can't let the graph/wikis grow unbounded, need really robust dedup and many AI-Consuela's

Someone is going to build a worldclass “Brain” for enterprises & make a stupid amount of money. Why? As @da_fant said, “coding w ai is solved bc all context is in the git repo. knowledge work is difficult bc context is spread out. an ai system that creates a git repo w all context for a knowledge worker will be able to 100% automate the work.” When companies talk about being data ready for AI, this is what they’re implicitly saying. Engineering has been prepared for this moment for a long time because of the deterministic nature of code, the centralization/versioning of data (read: GitHub), and AI tools that are largely build by engineers for engineers. But for the rest of white collar work, there’s a TON of catching up to do to properly harness the power of the technology. The big challenge here, and why no one has truly cracked the code for "an ai system that creates a git repo w all context for a knowledge worker" is because unlike code, most knowledge is 1) distributed, 2) unstructured, and 3) unverifiable. It's distributed: transcripts live in Granola. Documents in Notion. Customer Data in Hubspot. ERP. Emails. Slack messages. Random spreadsheets. SOP docs. Etc. Etc. Building an ingestion engine that connects to all of your disparate data sources and auto-updates based on the shelf-life of the data is the first, and frankly, easiest step of the process. Next, it's unstructured: let's say I want to create a proposal for a potential client. To nail the proposal, I want it to pull important information from a variety of sources. The specific asks & background from our initial sales call. Previous proposals to anchor ourselves to a proven format. And completed sprint boards from Linear, so the pricing & timeline in the document is grounded in truth. Whether it's a thoughtful filesystem (a la Obsidian) or an OpenClaw-esque memory structure, the brain needs to be great at self-organizing in a thoughtful schema. This is very hard, especially if you want to build a generalizable brain that can be shaped to an array of different enterprises. And finally, most knowledge is unverifiable: writing a function, running a unit test, and seeing if the code works is easy. It works or it doesn't. Using AI to accelerate your content creation process is highly subjective. What is a good/bad idea? Is the content in your voice or not? Does it feel like slop or novel? Answering these questions are both difficult and non-verifiable. That same system described above doesn't just have to be great at organizing & forming coherent relationships, but it also has to be great at self-improving based on feedback from the user. Memory systems (like those introduced by OpenClaw) are great to a point, but as you scale the corpus of data within your company's brain, things like compaction and cleaning become wildly important to avoid the needle in the haystack problem. Someone is going to figure out how to solve this problem, and when they do, not only will they make a shit ton of money, but they'll be robinhood for knowledge workers, enabling non-engineers to enjoy the sort of leverage that only technical folks have felt for the last few years.

@da_fant @businessbarista

@businessbarista x.com/da_fant/status…

@grixate excited to see what you end up building and how this would come together. I've also been experimenting with extending the karpathy method—linking all your wikis together to create a genuine knowledge network...check it out! - x.com/real_nac/statu…

hmm...

@shannholmberg cmux >>> youtube.com/watch?v=EUE8N6…

@real_nac cmux vs ghostty?

I use these daily, what would you add?

@thedevchandra building an open source llm knowledge network tool ! x.com/real_nac/statu…

I'm tired of seeing the same faces on my feed man I need brothers in tech, AI, startups, marketing, distribution, vibecoding, sf or elsewhere to build along like, drop your startup, let's make some friends.

@delveroin open source tool for creating a second-brain ! x.com/real_nac/statu…

Founders!! What are you building currently? Drop URL link! Let’s send some traffic there!!

@netcapgirl but did the slop really begin with ai?

it’s happening

@zuess05 the thing about slop is that even if the stuff you build today is straight slop, there will come a day when the models no longer just produce slop and then you can slop over your old slop and then, suddenly, it won't be slop anymore. it will just be good.

Vibe coding is just a dopamine addiction disguised as a business model. You type a prompt. You get a full app. You feel like a god for 10 minutes. Then you remember you are terrified of marketing. So you just open a new tab and build another app instead. The internet is about to become a massive graveyard of 24-hour side projects.

@CrewChiefBased great point.. you might find the knowledge network tool i've been building interesting x.com/real_nac/statu…

Karpathy released a guide on building a knowledge base that took off. Pair a knowledge base with what you need agents to do and strong results follow. This goes back to fundamentals - context. The more information provided the better an LLM can predict what’s next

@garrytan @chrysb trying to solve this with arkeon x.com/real_nac/statu…

@chrysb Trying to solve this with GBrain github.com/garrytan/gbrain

@bbhushanmishra demo is here as well: x.com/real_nac/statu…

@real_nac Interesting will try it out.

Karpathy drops one LLM Wiki gist → thousands of people instantly build their own. Now the builders need their own wiki 😂 We need a **Karapathy Wiki** — a public meta knowledge base for everyone hacking on LLM Wikis. Track: • Tool stacks (Obsidian + Claude, Cursor, local, GBrain…) • Best schemas & prompts • v2 lessons & what actually compounds vs rots • Who’s shipping what Who’s already building one? Drop your setup or repo below 👇 #LLMWiki #Karpathy

really interesting work. using symlink to store graph edges is quite clever. Curious how it traversal performs at scale, especially multi-hop traversal, path finding etc. ... i'm also curious how this system handles multi-agent writes + permissions, especially merging operations... by the way i'm working on a similar tool but it's postgres-backed (but still all local). would really value your feedback: x.com/real_nac/statu…