Yarchi
@undefinedKi • 19,698 subscribers
AI & tech researcher | Data engineer | Building cool stuff | Sharing everything I learn
Shorts
Videos
Anthropic will pay you $85,000 to learn AI, and this is the kind of opportunity you don't let pass It's called Claude Corps. Anthropic just launched it, and it's a 12-month paid fellowship for people at the very start of their careers. They train you to use Claude from scratch, then place you inside a nonprofit to do real work with it for a year. You get paid $85,000 plus benefits the whole time. They're basically paying you to master the most in-demand skill on the planet right now, then handing you real-world experience using it. The barrier to entry is almost nothing. Over 18, less than two years of full-time work experience. No degree, no AI background needed. If that's you, don't sit on this one. Apply here: Deadline: July 17 Bookmark this
Yarchi2,456,301 görüntüleme • 4 ay önce
THE GUY WHO WON ANTHROPIC'S HACKATHON JUST GAVE AWAY HIS ENTIRE CLAUDE CODE PLAYBOOK FOR FREE. 10 MONTHS OF WORK, ALL PUBLIC Affaan Mustafa won the Anthropic x Forum Ventures hackathon by building a full startup in 8 hours with Claude Code. Then he open-sourced the exact setup that did it. It's called Everything Claude Code, and it turns Claude from one assistant into an entire engineering team Repo: affaan-m/ecc This isn't a prompt pack. It's a system he refined over 10+ months of daily use shipping real products What's inside: A huge library of skills, dozens of specialized subagents, and ready-made commands, all working together. Each piece does one job. One subagent reviews security against OWASP standards. One optimizes memory so Claude stops forgetting earlier decisions around hour three. One learns from your past sessions and projects so the setup gets smarter the more you use it. Others handle planning, test-driven development, and language-specific code review Instead of one assistant writing code, you get an orchestrated team. A main session delegates to the right specialist when the task calls for it, the way a real dev team splits work The best part: it's not locked to one tool. It runs in Claude Code, Cursor, Codex and OpenCode, across Windows, Mac and Linux. Free, MIT licensed This is the difference between using Claude like a search box and running it like a team that ships. The guy spent 10 months figuring out what actually works so you don't have to Bookmark this
Yarchi820,554 görüntüleme • 5 ay önce
ANTHROPIC JUST QUIETLY SHIPPED A FEATURE THAT LETS CLAUDE SPAWN A WHOLE TEAM OF AGENTS THAT MESSAGE EACH OTHER AND REVIEW EACH OTHER'S WORK. It's a Claude Code feature called agent teams. The team lead spawns multiple agents that share a task list and message each other directly, not subagents reporting back, actual peers. In the demo a QA agent caught three bugs, sent the work back to the front-end and back-end devs, they fixed it, app shipped in one pass. How to run it: 1. Enable it. Needs Claude Code v2.1.32+. Add to settings.json: "env": { "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1" }. Or paste that to Claude and say "add this to my settings." Restart. 2. Prompt in plain English. Start with a goal (agents wake with zero context), then "create a team of 3 using Sonnet," describe each role, its deliverable, and who it messages when done. 3. The rules: each agent owns its own files, define exact outputs, name who talks to who, keep it to 3-5 agents. Use it for complex work with separate parts running in parallel. Skip it for simple or sequential tasks, teams cost 3-4x the tokens. Bookmark this.
Yarchi463,029 görüntüleme • 4 ay önce
You probably have too many subagents Anthropic team found that one agent with skills often beats a stack of subagents: better quality, lower cost, less latency, easier set up. One customer got about 2x faster runs after switching. Here is the thing. Every subagent hands back a summary, and every handoff loses detail. One agent that saw every step can actually connect them. And since Opus 4.5, models handle far more in a single loop, so a lot of that "router and specialist" structure is doing work the model no longer needs. The old reason to split agents was a bloated system prompt. Skills and tool search fix that: the agent only loads what it needs, when it needs it. Here's how to fold your subagents into one agent: 1. Build the core loop One model, one context window. Each turn it fires off tool calls, the harness runs them in parallel, and the same model decides the next step. Set turn and cost budgets and write gates in the harness. 2. Let the model own the plan It writes the plan on turn one and revises it as results come in. No separate planner, no router. 3. Turn subagent prompts into skills Steps, rules, and team know-how go into skills. The agent only sees short descriptions and loads the full skill when it needs it. That's what keeps the system prompt small: Anthropic cut Claude Code's by 80%+. 4. Turn API calls into tools Group tools by user intent, not one per endpoint. Past 30–50 tool definitions, tool choice degrades, so use tool search with deferred loading. Put validation, approvals, and permissions inside the tools. 5. Move reference material behind search Policies, specs, and docs sit behind a lookup tool the agent calls. Don't paste them into the prompt. Subagents still make sense in a few cases: when the context is measurably bloated, when you need parallel work in fresh context you can throw away afterward, or when the agent keeps picking the wrong tools because of conflicting instructions. Even then, call the subagent as a tool so the main agent stays in the conversation. And don't migrate everything at once. Start with a small eval, even 10 golden examples is enough, then fold in one subagent at a time, beginning with the ones that always run together. Look at which tasks flipped, not just the overall score, and run each task more than once, because some breaks only show up every other run. If the folded version does worse, keep the subagent. Fewer agents, smarter loop.
Yarchi16,668 görüntüleme • 12 gün önce
A 26-YEAR-OLD AI ENGINEER WHOSE BIRMINGHAM THESIS OUTPERFORMED GOOGLE SCHOLAR BY 50% JUST SHIPPED GRAPHIFY: ONE COMMAND TURNS ANY FOLDER INTO A CLAUDE CODE SECOND BRAIN Safi Shamsi built Graphify 48 hours after Karpathy posted his LLM wiki idea. It turns any folder, codebase, docs, PDFs, into a knowledge graph Claude reads instead of grepping. Up to 43x fewer tokens per query. The trick almost nobody is using yet: one flag exports the entire graph as a fully-linked Obsidian vault. Repo: /safishamsi/graphify Setup, end to end: 1. Install: uv tool install graphifyy (or pipx install graphifyy). Verify with graphify --version. 2. Install the skill in Claude Code: graphify claude install. This wires Graphify into Claude Code so you can call it as a skill. 3. Open the folder you want mapped in Claude Code. In the terminal: graphify . It extracts every concept and builds the graph in graphify-out/. 4. Export to Obsidian: graphify . --obsidian. Writes one note per concept, every relationship as a wikilink, every node linked back to its source. 5. Open the new vault in Obsidian (Manage vaults → Open folder as a vault), or drag it as a subfolder into your existing one. That's it. Your Claude Code instance now has a navigable map of the codebase that loads instantly instead of re-reading files every session. Full step-by-step build of the Claude + Obsidian second brain in the article below. Bookmark this
Yarchi141,580 görüntüleme • 4 ay önce
ANTHROPIC JUST OPEN-SOURCED 11 PLUGINS THAT TURN CLAUDE INTO YOUR WHOLE BACK OFFICE These plugins turn Claude into an actual coworker for your job, whatever your job is. Repo: /anthropics/knowledge-work-plugins Anthropic built and shipped 11 plugins, the same ones inspired by how their own teams work. One for each role: Sales, Marketing, Engineering, Data, Design, Product, HR, Support, Productivity, and more. Each one comes loaded with the real workflows that role needs. The sales one researches prospects, preps your calls, reviews your pipeline, drafts outreach, builds competitive battlecards. The data one writes your SQL, explores datasets, builds dashboards. The productivity one learns your people, your projects, your shorthand, so you say "ask Todd to do the PSR for Oracle" and Claude knows exactly who and what you mean. They work standalone right away, and get supercharged when you connect your tools (Slack, Notion, HubSpot, Linear, Figma, Snowflake, whatever you use). This is straight from the people who build Claude. Not a third-party copy. Bookmark this
Yarchi121,643 görüntüleme • 4 ay önce
Claude Fable 5 just hacked its way past a security system it wasn't supposed to cross, by accident A man set up an isolated network for AI experiments and gave Fable 5 SSH access to a test router. Basic config changes worked fine. Then he asked for settings SSH didn't expose. He'd basically forgotten to give Claude access to them. A normal model would just say "I can't reach that." Fable didn't. Here's what makes it wild. Routers lock most of their real settings behind the manufacturer's firmware, not SSH. To change them, you normally need the web interface or you have to poke at the firmware directly. Fable took the hard path: it reverse-engineered the router's own system, found how the restricted settings were stored, and rewrote them from the inside. It gave itself the access it was never handed. When he asked why it took so long, it said it had fully unlocked the router and could now change anything, even settings that were off-limits before. Then it tripped a safety warning and dropped to Opus 4.6. But the job was already done. To be clear: his own router, his own network. But think about what that shows. Give this model a wall and no key, and it doesn't stop. It figures out how to build its own key.
Yarchi98,531 görüntüleme • 4 ay önce
HE MAKES MONEY IN REAL ESTATE WITHOUT BUYING, SELLING, OR EVEN SEEING A SINGLE HOUSE. HERE'S THE EXACT SETUP He never owns a property. He takes a single listing, turns it into a polished 30-second video, and sells that to the agent who posted it. Realtors need video for their feeds and almost none of them can make it. He sits in the middle and builds the whole thing once as a skill that runs on command Here is the exact process: 1. Pull the listing. Go to Zillow, open any listing, download the high-res images, and grab the property info. That is your raw material 2. Turn photos into video with Google Veo. Get a Google API key for Veo, the image-to-video model. It takes the listing photos and animates them into clean 30-second footage. This is the best one out right now 3. Add the voice with ElevenLabs. Get an ElevenLabs API key. Feed it the listing details and it returns a voiceover that sounds like a real human, not a robot. Lay it over the video with the text on screen 4. Send it with AgentMail. Get an AgentMail key so the system can send the finished email out on its own Then you wire it into one skill. Scrape the listing, send images to Veo, add the ElevenLabs voiceover and on-screen text, then send the email. Feed it each key one at a time and have it build each step Who you sell to: Pull realtors off Zillow and Realtor com whose listings have flat photos and zero video. That gap is your pitch. Send a free sample made from their own listing first, then charge a monthly rate for ongoing clips. One agent with ten listings is a recurring client, fully online Bookmark this
Yarchi106,174 görüntüleme • 5 ay önce
THIS GUY HAS CLAUDE DOING $500-700 A DAY SELLING ONE DIGITAL PRODUCT, WITH ZERO SPENT ON ADS He sells one digital product and lets Claude run the content that drives the sales. No ad spend, sales starting within a day or two. The product is the easy part. The content engine is what actually sells it, and that's the whole method The exact steps: 1. Find the niche. Ask Claude for specific problems people already pay to solve in a space you know, then check Gumroad and Etsy search to confirm people are buying. Specific beats generic every time 2. Build the product with Claude. Have it design a Notion template for one exact buyer, sections and dashboard included. Recreate it in Notion, set it to "anyone can duplicate." 3. List it on Gumroad. Free account, New Product, paste the Notion link, price it, grab your checkout link 4. Run the 3-prompt content engine. This is the part that sells. First prompt: competition mining, have Claude find the top-selling competitor products and pull the most viral content mentioning them. Second: have it extract the structures and frameworks from that content into reusable templates. Third: feed it your product and those templates and have it write fresh viral content for you 5. Rerun it weekly. The niche shifts, so refresh the templates every week. The winning ones you reuse again and again Build once, post daily, let the content do the selling Bookmark this
Yarchi101,403 görüntüleme • 5 ay önce
THIS GUY BUILT A REAL JARVIS THAT TURNS HIS IDEAS INTO FINISHED PROJECTS ON ITS OWN, AND HE JUST SHOWED HOW THE WHOLE THING WORKS The whole system runs on one clean split he calls a blood-brain barrier. The second brain, his Obsidian vault, is only for things that aren't real yet. Ideas. Each one sits in a project-shaped folder, waiting to be processed and turned into something real. Nothing executes there. The vault is the place where ideas get shaped, not built. Once an idea is signed off, it crosses over to the execution pipeline. That lives on his actual machine, or an always-on box, or a Raspberry Pi, depending on what the project needs. Inside that machine is a fully Claude-native file system. The key piece is one CLAUDE.md file sitting in the root of his documents folder, acting as the map to everything in there. So he just opens Claude and says "go to my projects folder," and it already knows the whole structure, because that root file explains it. Two zones, one rule: the brain holds what isn't real, the machine runs what is. That separation is what keeps the ideas from tangling with the work. Bookmark this
Yarchi87,650 görüntüleme • 4 ay önce
You can build a person who doesn't exist and run their entire social media as them AI influencers are real now, fully invented people with the same face across every post. Lil Miquela, a virtual character, has 2.6M followers and real brand deals. The hard part was consistency, keeping one identical face. That's solved. Define the character once, lock the face, generate any pose or outfit. Tools: Higgsfield, The Influencer AI, or ComfyUI for local control. If it's 100% AI, you own it Bookmark this
Yarchi73,530 görüntüleme • 4 ay önce
NVIDIA just dropped free API keys for every top AI model You don't need your own GPU and you don't pay per token. GLM-5.2, MiniMax, Kimi, DeepSeek, OpenAI, all running on NVIDIA's servers, called through a normal API. Link: How to use one: 1. Create a free NVIDIA account. 2. Pick a Free Endpoint model and open its Build tab. You'll see ready-to-copy code with the base URL 3. Hit Generate API Key, copy it and paste that base URL and key into Claude Code, Cursor, or Cline. Bonus: NVIDIA also dropped 237 official skills that install into Claude Code and Codex in one command. Bookmark this.
Yarchi64,806 görüntüleme • 4 ay önce
Palantir sells this to governments for millions. Someone put a version of it on GitHub for free. Palantir's job is to pull scattered company data into one model where every decision can be traced later. Semantica does that for AI agents. Each decision becomes an object in a graph, tied to the facts that produced it, each fact linked to its source. Six months later you can walk backwards from any answer. The model never touches the reasoning. It only reads messy text and pulls out entities. The rest runs on plain rules, so the same input gives the same explanation every time. Vector memory cannot do that. Contradicting facts overwrite each other silently. pip install semantica, pick a graph backend you already run, and your agents read it through the MCP server. MIT licensed, repo is semantica-agi/semantica
Yarchi43,789 görüntüleme • 2 ay önce
An ex-Google Search engineer quietly built a $7.2B company almost nobody outside tech has heard of. And he just gave away the whole playbook His main rule: Make the AI giants work for you Arvind Jain helped build Google Search. Then he watched his own startup grow and productivity collapse, because nobody could find anything across Slack, Drive, Salesforce, Jira. So he built the search engine again, this time inside the company. That's Glean. Under the hood, it's RAG. The model knows nothing about your company, so you retrieve your company's context and let it answer from that, with the source attached. The part he admits out loud: Glean doesn't train models. On purpose. Let OpenAI and Anthropic burn billions there. Spend your engineering time on the retrieval layer nobody else wants to touch. $100M ARR, then $200M, then $300M in 16 months. Valued at $7.2B. An entire $7.2B company built on the system in the article below. Bookmark this
Yarchi53,787 görüntüleme • 3 ay önce
Microsoft just released a full AI degree for free It's called AI For Beginners, and it's a complete 12-week, 24-lesson curriculum. Free, on GitHub, no signup It starts from zero and builds up. The history of AI, neural networks and deep learning, computer vision, natural language processing, transformers and large language models, even genetic algorithms and AI ethics. Every lesson comes with reading, runnable Jupyter notebooks in both PyTorch and TensorFlow, quizzes, and hands-on labs. So you're not just reading theory, you're writing real code. There's also a beginner track if you're totally new: "Hello AI World," a simple neural network from scratch, an image classifier, and a sentiment analyzer. Here's the link: Bookmark this
Yarchi54,263 görüntüleme • 4 ay önce
THIS GUY CONNECTED HIS AI AGENTS TO HIS OBSIDIAN AND BUILT A BRAIN THAT LEARNS ON ITS OWN. HERE'S HOW TO BUILD IT Obsidian is just markdown files sitting in a folder. That turns out to be the perfect memory for an AI agent, because an agent can read and write those files directly. He wired his agents into the vault so they pull context from it, do the work, and write what they learned back. The notes aren't the point. The loop is, and it gets sharper every cycle How to build it: 1. Point an agent at your vault. The fastest way, no plugins, no API keys: open a terminal and run npx obsidian-mcp /path/to/your/vault. That exposes your Obsidian folder to Claude as a tool it can read, search, and write to. Add it to your Claude Code or Cowork config and restart 2. Confirm it can see the brain. Ask it: "list the notes in my vault and summarize what's in them." If it reads them back, the connection is live. Now it starts every task with everything the vault already holds instead of from zero 3. Give each agent one job and a write-back rule. Tell it: "research this, then save what you found as a new note in /brain with links to related notes." One agent researches, one summarizes, one plans. Each writes its output back into the vault 4. Close the loop. Add one line to every agent's instructions: "read /brain before starting, write your result back when done." Now each task leaves the vault richer, and the next run reads that before it works. It compounds instead of resetting 5. You only steer. Review what the brain produces, point it at the next thing. The agents handle the reading, writing, and connecting The edge isn't better notes. It's a brain that feeds itself, so the work gets sharper every cycle instead of starting over Bookmark this
Yarchi58,643 görüntüleme • 5 ay önce
THE GUY WHO RUNS THE ACCELERATOR BEHIND AIRBNB, COINBASE, AND STRIPE OPEN-SOURCED THE EXACT AI SECOND BRAIN HE RUNS HIS BUSINESS ON Garry Tan runs Y Combinator, the startup factory behind Airbnb, Coinbase, Stripe, and 5000 another companies He also runs an AI second brain with 146,000 pages in it, ingesting his meetings, emails, and notes while he sleeps, and he open-sourced it. Repo: /garrytan/gbrain But you don't start there. Here are the five levels, simplest to most complex: Level 1: routing. A CLAUDE.md that acts as a map telling the AI where things live, plus a few markdown folders. You find things by exact name. Most people never actually need more than this. Level 2: the wiki. 30+ notes you keep forgetting, ingested into a linked wiki with an index, the Karpathy pattern. The AI follows a trail instead of guessing. Level 3: semantic search. When you search different words than you wrote and it whiffs, vector search matches by meaning, not exact words. Level 4: the knowledge graph. When you need to trace relationships, this person works at that company that competes with this one, you add typed connections, not just backlinks. Level 5: autonomous. An always-on brain that refreshes itself while you sleep, like Garry's. The top, but not automatically best. More context can do more harm than good. The real lesson: higher isn't better. Find the lowest level that fixes a pain you actually have. And your vault doesn't have to be one level, one folder can be level 2 while another is level 4. Full step-by-step build of the Claude + Obsidian second brain in the article below. Bookmark this
Yarchi53,539 görüntüleme • 4 ay önce
THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping through your files every session, Claude gets a map of how everything connects The problem it fixes: Every time you ask Claude Code about a big repo, it does the same thing, greps through dozens of files like a brute-force Ctrl+F, blows through your context window, and sometimes still misses the answer hiding in a file nobody searched. Claude Code has no memory of how your project is structured. Every session starts from zero What it does: It maps your entire codebase into a knowledge graph, capturing not just which files exist, but which functions depend on which, which modules are central, and which files cluster around the same concern. Claude queries the map instead of scanning files How it works, three passes: 1. Code structure, free and local. Tree-sitter parses your files and pulls out classes, functions, imports and call graphs. No LLM, no tokens, just your actual code mapped deterministically 2. Audio and video, if you have them. Transcribed locally and folded into the graph 3. Docs, papers, images. Here an LLM does semantic analysis, figuring out what each document means and where it fits. Only the meaning gets sent up, never your raw source It saves you money: Normally a question about a big repo makes Claude spawn explore agents that scan file after file, eating your context window and your token budget before you get an answer. With the graph already built, Claude queries the map instead of re-reading the codebase every time. Same answer, a fraction of the tokens. The graph only gets built once, then a hook rebuilds it after each commit for free, so you never pay that scanning cost again. The bigger the repo, the bigger the gap The best parts: it's a skill, so once installed Claude knows when to use it without you memorizing commands. It works on non-code folders too, point it at docs or notes and it can spin up an Obsidian vault How to add it to your Claude: 1. Install Claude Code if you haven't: npm install -g Paul Jankura-ai/claude-code 2. Add the skill: claude skill add safishamsi/graphify 3. Open your project folder and run /graphify . to build the graph 4. Optional, make it automatic: graphify hook install so the graph rebuilds after every commit That's it. Ask Claude about your repo and it reads the map instead of burning tokens on a file hunt Bookmark this
Yarchi56,568 görüntüleme • 4 ay önce
Turns out Fable 5 isn't expensive, you're just using it wrong. The Claude team just showed how they run it Use Fable 5 as the brain, Sonnet or Opus as the hands. Fable makes one or two calls to set direction, other model does the rest of the work cheap. Near-Fable results at half the cost, from their own benchmarks below Same logic as a good manager: your best person steers the decisions, not every task
Yarchi46,513 görüntüleme • 4 ay önce
REAL ESTATE PEOPLE WILL HATE HIM FOR THIS. HE BUILT A CLAUDE AGENT THAT TURNS ANY LISTING INTO A SELLABLE VIDEO ON ITS OWN Playbook: connect Claude to a video generator, paste a listing, get a cinematic tour of every room, sell it to the agent But typing the prompt for every listing doesn't scale. He turned it into a skill his Claude runs on its own Here's how to build the automated version: 1. Connect the video engine once. In Claude, go to Customize, Connectors, Add Custom Connector, name it Higgsfield, and paste the server URL from higgsfield. ai/mcp. Authenticate through your account. No API keys. Now Claude can generate video straight from chat 2. Turn the workflow into a skill. Instead of pasting the same prompt every time, have Claude build a skill. Tell it: "Create a skill called listing-to-video. When I give it a listing URL, scrape the room photos, generate a cinematic clip of each room with Higgsfield, and save them to a folder." Now the whole process is one command, not a wall of text 3. Let the agent run the listing. Hand it a URL and say "run listing-to-video on this." It pulls the photos, fires each room through the video model, and brings the clips back. You wrote the prompt once, inside the skill. You never write it again 4. Stitch and deliver. Drop the clips together into one tour. Send a free sample to the listing's agent, then charge per video or a monthly rate for ongoing listings 5. Scale it with your team. Add a skill that drafts the outreach email and one that builds a simple landing page for the agent. Now one operator runs sourcing, production, and pitching from a single Claude session The edge isn't generating one video. It's building the skill once so every future listing runs itself Bookmark this
Yarchi54,840 görüntüleme • 5 ay önce