King Chatterjee @mindstatic

Learn AI for free directly from top companies. 1 - Anthropic: anthropic.skilljar.com 2 - Google: grow.google/ai 3 - Meta: ai.meta.com/resources/ 4 - NVIDIA: developer.nvidia.com/cuda 5 - Microsoft: learn.microsoft.com/en-us/training/ 6 - OpenAI: academy.openai.com 7 - IBM: skillsbuild.org 8 - AWS: skillbuilder.aws 9 - DeepLearning.AI: deeplearning.ai 10 - Hugging Face: huggingface.co/learn Comment "Learning" if you find this helpful. Repost so others can take help. Must bookmark for future reference.

Kimi K3 can now be run locally! ✨ The 1-bit model retains ~78.9% accuracy after we shrunk it from 1.56TB to 594GB (-62% size). Run on a Mac Studio + 128GB RAM device. Kimi K3 is the strongest open model to date. Guide: unsloth.ai/docs/models/ki… GGUF: huggingface.co/unsloth/Kimi-K…

Kimi.ai @Kimi_Moonshot

3 weeks ago

Introducing Kimi K3: Open Frontier Intelligence 🔹 2.8 Trillion Parameters, 1 Million Context, Native Multimodal 🔹 Kimi Delta Attention enables up to 6.3x faster decoding in million-token contexts 🔹 Attention Residuals deliver ~25% higher training efficiency at <2% additional

keyboard_arrow_left Previous keyboard_arrow_right Next

2K 8K 58K 25.3M 23K

every founder in SF and NYC should read this before you start a company: Steve Jobs' building philosophy.

ANTHROPIC'S LEAD ENGINEER WON A $1.2M BONUS FOR A SYSTEM THAT TURNS ANY DATA CHAOS INTO A GRAPH IN 8 STEPS raw chaos in - self-updating graph out - and the agent gets +42% productivity from day one Load → Extract → Graph → Index → Query → Memory → Swarm → Update eight steps, one pipeline, graph grows while you sleep documents, code, Slack - everything into one stream through Claude Code - nothing gets lost Fable 5 + Opus 5 extracts entity relationships - Neo4j builds a live structure - zero duplicates three types of search in one answer - vectors, keywords and graph - merged ranking gives accuracy no RAG delivers alone a nightly agent pulls new data and updates the graph automatically - the system gets smarter while you sleep bookmark and paste into Claude Code - a $1.2M system now free

Sprytix @Sprytixl

3 weeks ago

x.com/i/article/2078…

27 102 768 1.3M 3K

Two Anthropic seniors just made Karpathy's loop 1000x better with "Graph Engineering" - dropped 11-page PDF the shift: the agentic systems got 1000x better the moment you wired agents into a graph here's the playbook in 6 steps: step 1 → build one loop: generate, critique, revise - one self-review cycle beats a smarter model with none step 2 → add tools: search, code execution, database - thinking without tools is hallucinating step 3 → go parallel: spin up agents in separate worktrees - same repo, different branches, no conflicts step 4 → add a graph: agents write findings as typed nodes and edges - not transcripts - every claim keeps its source step 5 → ground your evaluator: it checks claims against graph edges, not vibes - "Triple not found" beats "seems off" step 6 → the graph survives every session - your agents stop rebuilding context from scratch the result: Karpathy ran 1 agent in 1 direction - this system runs 1,000 with shared memory - same model, it's the architecture read this 11-page PDF and paste it into your Claude - you won't regret it bookmark - then read the article on building graphs from scratch ↓

codila @0xCodila

3 weeks ago

x.com/i/article/2079…

37 266 2K 3.5M 5K

I think the most interesting thing about Jack Dorsey's "Slack killer" is the idea around shared compute. I haven't seen people talk about it so here are my thoughts FWIW: Open models got good, close enough to the paid frontier stuff to run for real. But the strongest ones need expensive hardware most people probably won't buy alone, and it's kinda a pain to set up if you aren't technical. Shared compute solves exactly that. In Buzz, one person runs the machine, loads up an open model like Google Gemma, and everyone in the community plugs into that same model. Basically, a whole group has real AI they own and control together, running on their own hardware, learning from their own data. Once you see it, a bunch of things click into place. 1. A community can now run a top open model together, on a machine they own, instead of renting from a lab. 2. It learns from the group's private data and gets sharper over time, and all of that stays inside the community. 3. A narrow, private model can quietly get better than ChatGPT for the one world your group lives in. 4. It's impossible to copy, because the edge is the private data on your machine, not the model itself. 5. The moat stops being how smart your AI is and becomes whose data it learned from. 6. Compute becomes something you share like a building shares a gym. 10 people split one machine instead of 10 people each renting forever. 7. Idle compute becomes income!!! Your machine sits dead half the day, so it earns money renting that time to someone who needs it. 8. Communities become the unit of intelligence instead of companies. The group with the smartest shared brain wins, and being a member means owning a piece of it. 9. A shared brain becomes an asset you build equity in. You put in money and data, it appreciates, and your slice is worth something the day you leave. 10. The whole thing runs on open protocols, so the group keeps full control and nobody outside can throttle it or shut it off. You know me, obviously, my head went to what startup ideas come to mind here. Adding them to @ideabrowser soon. Well… 1. The vertical brain. Pick one profession, tax lawyers or real estate agents or indie game devs, and build the shared machine trained on everything that group knows between them. A year in it's the smartest AI in that field, impossible to copy, and you own the club it lives in. 2. The rental marketplace for collective brains. Once these private models exist, outsiders will pay to use them. You build the layer where a group lists its brain, an outsider pays per task, and the money flows back to the members while you take a cut. A marketplace for expertise, not compute. 3. The idle-compute exchange. Every shared machine sits unused half the day. You build the market that rents that dead time to whoever needs the power right then, so owners earn money off a machine that was just sitting there. Idk where Buzz goes, but it's cool to see Jack putting it out. Right now the way it works in AI is you rent your intelligence from a few giant labs that own the machine, set the price, and hold the off switch. Shared compute flips that, because a community can run the model together, feed it their own private data, and keep full control of the whole thing. It's one of those things that might look tiny today, but Jack does has a habit of being early.

Vinny @hot_town

2 weeks ago

I tried @jack's Buzz. It's like Slack + OpenClaw + Herdr + but with some really unique features that people are sleeping on. The video below shows how it works, and some of my thoughts on the process and platform, e.g.: - Create and interact with agents on top of any harness

132 196 3K 1.3M 5K

So i didn’t really know what graph engineering is, and i still don’t really… but it’s basically just langgraph?

Shann³ @shannholmberg

3 weeks ago

what's the difference between a loop and a graph? (marketing edition) both are ways to run an agent, the difference is who decides the path, the agent or you. a loop still starts with you. you set the goal, the brief, and the bar it has to clear. what the agent owns is the

72 215 2K 285K 2K

x.com/i/article/2064…

Andrew Ng just released a 1-hour course on building agentic knowledge Graphs from scratch: • 00:00 - Introduction to agentic knowledge Graphs • 03:07 - Construction of agentic Graphs • 14:00 - Architecture of multi-agent systems • 23:00 - Building agentic graphs with Google ADK • 01:06:03 - Why Graphsare the future of agentic AI Worth more than 10 articles on loop engineering. Watch it today, then read how to become a graph engineer in the article below.

Codez @0xCodez

3 weeks ago

x.com/i/article/2079…

110 708 4K 5.7M 13K

CHINA JUST KILLED THE OCR BUSINESS. A 3B-parameter model the size of a peanut can read an entire 100-page PDF in one shot. No page splitting. No context loss. No cloud bill. Meet Unlimited-OCR 👇 • Reads full documents with a 32K context window • 93% on standard OCR parsing benchmarks (+6 over baseline) • Error rate stays below 0.11 even after 40+ pages • Multilingual out of the box • Runs 100% locally on your hardware • Supports Transformers, vLLM, SGLang, Docker, Ollama & llama.cpp Here's the crazy part: Most OCR tools still process documents one page at a time. Unlimited-OCR reads the entire document as a single context, preserving tables, references, layouts, and cross-page relationships. That changes everything. Meanwhile companies are paying: • $1.50–$15 per 1,000 pages • Sending sensitive PDFs to cloud providers • Waiting for API responses This model does it offline. For free. Forever. Built by Baidu to push beyond DeepSeek-OCR. Already 1.9M+ downloads on Hugging Face... ...and almost nobody is talking about it yet. Open source is moving faster than most enterprise software.

The "harness" is starting to blur with the neural architecture, in terms of who carries the inductive biases that unlock generalization. We show that training RLMs specifically is far superior at scaling and generalization to harder tasks than training vanilla Transformers.

alex zhang @a1zhang

3 weeks ago

Transformers struggle to generalize to tasks they were not explicitly trained on. Instead, we propose in 2026 that it is the job of the harness to generalize through composition. We observe a powerful property when training RLMs: for tasks with shared structure that look

73 386 3K 836K 3K

"If—" is a poem by English poet Rudyard Kipling If you can keep your head when all about you Are losing theirs and blaming it on you; If you can trust yourself when all men doubt you, But make allowance for their doubting too: If you can wait and not be tired by waiting, Or being lied about, don't deal in lies, Or being hated don't give way to hating, And yet don't look too good, nor talk too wise; If you can dream—and not make dreams your master; If you can think—and not make thoughts your aim, If you can meet with Triumph and Disaster And treat those two impostors just the same: If you can bear to hear the truth you've spoken Twisted by knaves to make a trap for fools, Or watch the things you gave your life to, broken, And stoop and build 'em up with worn-out tools; If you can make one heap of all your winnings And risk it on one turn of pitch-and-toss, And lose, and start again at your beginnings ⁠And never breathe a word about your loss: If you can force your heart and nerve and sinew To serve your turn long after they are gone, And so hold on when there is nothing in you ⁠Except the Will which says to them: 'Hold on!' If you can talk with crowds and keep your virtue, Or walk with Kings—nor lose the common touch, If neither foes nor loving friends can hurt you, If all men count with you, but none too much: If you can fill the unforgiving minute With sixty seconds' worth of distance run, Yours is the Earth and everything that's in it, ⁠And—which is more—you'll be a Man, my son!

Microsoft has released an open-source tool that helps teams learn ontology design before choosing a knowledge graph platform. It is called Ontology Playground. The project is a fully static React app, which means it does not need a backend, account system, database, or hosted service to run. The goal is simple: Help people understand what goes inside a graph before they buy or build the graph database. Ontology Playground includes six pre-built domain ontologies: → Retail → Healthcare → Finance → Manufacturing → E-Commerce → Education Each one gives users a starting point for understanding entities, relationships, properties, and how domain knowledge gets structured. The app also includes a live visual designer, structured learning paths, hands-on labs, and RDF/XML export for Fabric IQ. Because it is static, it can be deployed almost anywhere. No backend. No vendor lock-in. No platform commitment upfront. This matters because many teams jump into knowledge graphs too early. They focus on the database first. But the harder question is usually: What should the graph actually know? Ontology Playground teaches that layer first.

Recommended reading from the legendary @piersonmarks

Pierson Marks @piersonmarks

a month ago

x.com/i/article/2075…

15 31 419 153K 2K

This is one of the best breakdowns on the fundamentals of LLMs I've ever read. Anytime someone asks me for resources to climb the steep AI learning curve, I always provide the same list. 1) @3blue1brown's neural network videos 2) @karpathy's zero to hero playlist 3) @dwarkesh_sp's whiteboard explainers Now @_raghavdixit_'s "Vectors are all you need" and future articles in the explainer series are getting added to the list.

Raghav Dixit @_raghavdixit_

a month ago

x.com/i/article/2074…

23 111 835 499K 3K

The approach here is exactly right. Pairing builders with domain experts, and grounding the work in the company’s real knowledge, systems, and workflow context. Most AI initiatives don't stall for a lack of ambition or model capability, but because the people building the tools or workflows are too far removed from the friction of the actual work. The breakthrough happens when you combine technical builders, domain experts, and the scattered knowledge, workarounds, context behind the workflow itself. You can’t redesign the future if you’re too far removed from the friction of the present.

Praveen Neppalli @praveenTweets

a month ago

Agentic AI adoption is on fire at @Uber, and it's changing the way we build, not just in engineering, but across the entire company. Today, 99% of our engineers use AI tools. More than 70% of pull requests are attributed to local or cloud agents. And our engineers have built

179 373 3K 1.7M 5K

China has killed the entire vector database industry. They open-sourced TencentDB Agent Memory. It gives any AI agent long-term memory that runs 100% locally. No Pinecone. No cloud APIs. No repeating yourself every session. - 61% fewer tokens - PersonaMem accuracy: 48% → 76% - Zero external API dependencies - Runs on plain SQLite Most memory systems compress your history into an opaque vector pile. when recall goes wrong, you're guessing. this one doesn't compress, it builds a semantic pyramid. L0 Conversation → L1 Atom → L2 Scenario → L3 Persona. Short-term state gets encoded as a Mermaid graph in your agent's context. verbose tool logs get offloaded to disk. when the agent needs proof, it drills back via node_id to the exact raw log. no lossy compression. every layer is readable markdown you can just open and inspect. 5.1k stars. 100% Open Source.

Many are asking how Replit is improving so rapidly—we closed the loop and the agent is self-improving. Technical details here:

Michele Catasta @pirroh

a month ago

x.com/i/article/2070…

31 128 1K 938K 4K

great post by @delba_oliveira

ClaudeDevs @ClaudeDevs

a month ago

x.com/i/article/2074…

330 2K 18K 6.2M 39K

Warm take: Your world model should never stop learning Introducing AdaJEPA, an adaptive WM that plans, acts, and adapts in a closed loop. Every action leads to a new observation, and every transition refines the latent representation and prediction. 📝: agenticlearning.ai/adajepa/