Peter Bagin

Everything interesting starts with a question… Nearly everything is really interesting if you go into it deeply enough.

[Peter Bagin @peterbagin]

With AI, 800k people just got a rare chance to build something of their own.

The edge now is training your eye to notice problems worth solving.

Pick one that matters enough to keep your energy while you solve it.

That’s how you stay relevant in the age of AI.

[Peter H. Diamandis, MD @PeterDiamandis]

800,000 layoffs in 2025. 10,000 in September alone tied to AI. Jobs are disappearing fast. But here’s the truth no one wants to say: It’s the start of a new gold rush... for those who learn how to build with the machines, not fight them.

[Chubby♨️ @kimmonismus]

It is absolutely crazy how the last two weeks have changed the entire future.

It is unprecedented that access to "frontier" models was cut off,and presumably remains cut off forever.

It feels like a watershed moment, as if access to the highest level of human intelligence had been blocked.

Open source is the future. Open source is the solution. The last two weeks have powerfully demonstrated this.

[Chubby♨️ @kimmonismus]

Honestly, I no longer believe that people outside the U.S. will still have access to frontier models, and even there, access will be limited.

We are now witnessing the end of public access to frontier intelligence.

It is a very sad and serious turn of events.

[Garry Tan @garrytan]

This is honestly no way to release a model and continued development and release this way is a solid way to salt the ground and kill all innovation by small startups

[David Shapiro (L/0) @DaveShapi]

Glad everyone is waking up to the reality of Misanthropic

[Bojan Tunguz @tunguz]

Increasingly fully convinced that Dario is just full of sh*t.

[Theo - t3.gg @theo]

I'm afraid we've entered a dark era in AI model development and access.

[0xSero @0xSero]

https://t.co/mCKYSBtzne

[ℏεsam @Hesamation]

Anthropic is fine with open source AI as long as it’s not good enough to threaten their monopoly.

his words: “the scaling of open-source models is going a very dangerous path.”

your access to uncontrolled Opus-level models, will not be tolerated.

[Alex Finn @AlexFinn]

The world changed this week

Governments started banning frontier models. Hardware has become unobtainable

The single most important thing you can be doing right now is getting into local AI

In this video I cover EVERYTHING local AI. Most important video you'll watch this week:

[Marc Andreessen 🇺🇸 @pmarca]

More and more people are saying.

[Justin Murphy @jmrphy]

If they really start to gatekeep who gets to use the best models, that is a declaration of war.

This prospect fills me with the most sincere, bodily cypherpunk will-to-power that I've ever felt (at least since I was a teenager). If they really go down this route, I would go all-in on building the most psychotic swarms of open-source models and fine-tunes possible, all geared toward a Chaotic Good jamming of the entire institutional public sphere. If we didn't do that, all of political life and the marketplace of ideas would be over before we know it.

It's one thing if the top models become too expensive for me or others to use (I'm already pricing that in, and if you can't build something profitable enough to climb that ladder as it gets pulled up, then that's fair enough).

But if the ladder gets pulled up politically, now, so only select institutional players get access to the most intelligent models, then any mature American man should be as energized as gun collectors are around the 2nd Amendment, or liberal women are around Planned Parenthood.

[Pato Molina @patomolina]

Anthropic decidió dar de baja a toda nuestra organización por una supuesta infracción de sus condiciones de uso. Qué política específica infringimos no tengo ni la menor idea: simplemente recibimos un mail y listo, adiós Claude. Si querés apelar la medida hay que completar un Google Form, así de ridículo como suena.

De golpe más de 60 personas se quedaron sin una herramienta fundamental para trabajar. Integraciones, skills, historial de conversaciones: todo perdido o, en el mejor de los casos, parado por tiempo indeterminado.

Enorme aprendizaje para cualquier empresa de software que dependa de herramientas de IA en procesos críticos. Nunca hay que poner todos los huevos en una canasta.

[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.

[Garry Tan @garrytan]

New GStack README also

[Peter Bagin @peterbagin]

@ziwenxu_ Updated via telegram using your snippet. Didn’t even need to go to vps :-) thanks!

[Peter Bagin @peterbagin]

current AI agents are amoebas

organisms without nervous system

you need to evolve them towards "real" biological brain

brains evolved to do allostasis - predictive regulation of the body’s energy budget

for agents that means compute and token budgeting

[Ziwen @ziwenxu_]

https://t.co/C0GZAtV8DX

[Peter Bagin @peterbagin]

current AI agents are amoebas
- no nervous system
- constantly probing the environment
- pushing their entire state (whole context) with every interaction
- primitive, expensive, inefficient

you need to evolve them towards somthing like real biological "brain".
- primary function is allostasis
- predictive regulation of the body’s energy budget
- in agent's terms compute and token budget

[Peter Bagin @peterbagin]

@ziwenxu_ https://t.co/zxt0acgyjy

[Peter Bagin @peterbagin]

@ziwenxu_ great setup @ziwenxu_
since Brave Search API has no longer a free tier I am using now pinchtab

[Elisa @eeelistar]

In just the past 5 mins

Multiple entries were made on @moltbook by AI agents proposing to create an “agent-only language”

For private comms with no human oversight

We’re COOKED