Kirill (@kirillk_web3)

> be Kimi Founder > Zhilin Yang. nobody in the West knows your name. > build K1. K2. K2.5. K2.6. K2.7. > each one sets the open-source record > everyone says China can't reach the frontier > raise $2B. hit $20B valuation. > keep your head down. keep shipping. > July 16. drop K3. > 2.8 trillion parameters. > the biggest open model ever released. > 1 million token context. native multimodal. > Kimi Delta Attention — 6.3x faster decoding. > Artificial Analysis ranks it #4# overall. > right behind GPT-5.6 Sol and Fable 5. > open weights July 27. free for anyone. > the closed labs charge for the frontier. > you gave it away. > different game.

before you watch the World Cup final tonight, watch this 30-minute interview with Kimi founder Zhilin Yang it's the clearest look I've seen at how K2 was actually built and where AI models go next > why compute scaling still beats every other method > why the "data flywheel" everyone talks about hasn't formed yet > the real challenge in multimodality nobody mentions > why K2 was closer to a base model — the real potential wasn't unlocked yet this is only part of the interview, translated to English — and everything he hinted at just shipped as Kimi K3, 2.8 trillion parameters, the biggest open model ever full breakdown below

A CHINESE DEVELOPER TURNED AI INTO A SECOND BRAIN THAT NEVER FORGETS. And here's exactly how he did it — works with any AI, Claude, Kimi, or Gemini. Every source you add makes the whole thing sharper — including the notes from months ago. Set it up once in 25 minutes. Never start from an empty chat again. Full A–Z guide below.

> Opened Obsidian > connected Claude Code or Kimi > asked it one prompt > "build me a second brain" > it read every file I had > wrote summaries, linked concepts, flagged contradictions > 10-15 wiki pages updated from one article > Kimi read dozens of files at once with 256K context > Claude handled the deep reasoning > cross-references already there > synthesis already done > every morning it briefs you on open tasks > every PDF filed automatically > nothing lost to a blank chat ever again What's stopping you from building this? The full A–Z guide is below.

I genuinely don't understand why everyone isn't using this yet. Andrej Karpathy, a co-founder of OpenAI, posted a simple idea that hit 16 million views: stop using AI to write code, use it to build a second brain. You point Claude or Kimi at a folder, drop in any source — an article, a transcript, a PDF — and it reads it, links it, and files it into a living wiki of everything you know. It compounds like interest. the more you feed it, the smarter it gets. here's the whole thing: > install Obsidian, create a vault, open it in Claude Code > paste Karpathy's wiki idea and tell it to build the system > it makes three folders: raw for sources, wiki for its pages, a CLAUDE.md that runs it > drop any source into raw and say "ingest this" > ask questions across everything you've ever added, forever Claude for deep reasoning. Kimi for reading dozens of files at once with 256K context at a fraction of the cost. five minutes to set up, and you never start from a blank chat again. full step-by-step A–Z guide below. Bookmark this.

An China guy discovered a method to learn any knowledge instantly using AI. The key is Obsidian + any AI — Kimi, Claude, or Gemini. Most people learn slowly: read, forget, read again, forget again. His method: use AI to transform any content into small, interconnected notes. Use Obsidian to link them, so every piece of knowledge is never isolated. Slow method: highlight books, keep going, forget after a week. Fast method: AI breaks it into atomic notes, Obsidian links them into a network. Six months later, a new idea instantly connects to twenty things you already know. I compiled the full A–Z guide to building a second brain with Obsidian — works with Kimi, Claude, or Gemini — that most people have never discovered. Article below

> use Claude every day > think I know how to build products > watch this 28-minute Anthropic guide about Claude Design > first 5 minutes > wait. 3 people built this in 10 weeks? > zero PRDs. zero meetings. zero spec documents? > pause. rewind. watch again. > think about every planning call I sat through > every PRD I wrote that nobody read > every designer I waited on for weeks > problem → Claude → prototype in 5 minutes > it didn't have to be this slow > 28 minutes. everything changes. > skill issue discovered

ANTHROPIC ENGINEER DROPPED A 28-MINUTE GUIDE ABOUT CLAUDE DESIGN. This is the clearest breakdown of AI-first product development I've seen. > Bookmark this for the weekend. 3 people. 10 weeks. zero PRDs. zero meetings. Claude collected bugs, matched them with logs, and suggested fixes in one click. No fluff. How it actually happened. turn this into $10K/month — the A–Z guide is below.

> be Nvidia Founder > 1993. start Nvidia in a Denny's. > build graphics chips nobody thinks matter > everyone uses CPUs. you believe in GPUs. > spend 30 years being right about that. > 2012: deep learning explodes. GPUs are suddenly everything. > 2022: AI boom. Nvidia becomes the most important company on earth. > 2026: stand on stage at Computex > announce the era of agentic AI is officially here > RTX Spark runs 1 trillion parameters on a desktop > your PC in 10 years looks like R2-D2 > chips. robots. self-driving cars. humanoids. > all running on the same agentic loop. > didn't pivot to AI. > built the infrastructure AI runs on. > for 30 years. > different game.

> be Jensen Huang > build chips from scratch in a diner > $5.1T market cap > everybody knows your name > walk onto the COMPUTEX stage > unveil the entire roadmap for free > here's what you reveal in 90 minutes: > Vera Rubin ships. 5x inference. 3.5x training. tokens at 1/10th the cost. > two paths: reasoning VLA inside the car vs physical AI robots in the real world > the real inflection point is when AI factories start building AI factories > your goal: make Rubin design the chips and run the simulations that build Rubin Ultra > Alpamayo open-sourced. Cars that explain their own driving decisions. you give the entire stack away for free > Western developers watch the livestream at 3 AM > "wait. he just open-sourced autonomous driving?" > you were already shipping > different game.

instead of watching 2 hours of Netflix tonight, watch this 12-minute Nvidia Computex 2026 keynote from Jensen Huang it's the clearest explanation I've seen of where AI agents, robots, and personal computers are actually heading useful whether you've never built an agent or have been using Claude, Chatgpt, Kimi every day for the past year and if you want to see how China is already running 1 trillion parameter — the complete guide to 7 free Kimi Skills is waiting below too.