Ronin
@DeRonin_ • 117,345 subscribers
20 / CEO of Arcane / Angel Investor / engineering virality for AI companies
Shorts
![Somebody released a directory of 1,700+ dead startups (backed by YC) here is the link: all of them have deep analysis of product inside and full funding story (in short, 100% product's description) many of these startups failed not because the idea was bad, but because they launched at the wrong time or were just grant cash grabs the ideas themselves are still worth drawing inspiration from (i've already found few ideas of my further startups btw) [ BOOKMARK ]](https://image.24vids.com/tw-2030941646414581921/amplify_video_thumb/2030941579763134464/img/Ufh63Sl6sUyrbztP.jpg)
Videos
everyone is using GPT-6 Astra to make games... i had it build me a robot prototyper i type what robot i want, and it builds the whole thing and tells me how to make it real one prompt: "prototype me a full Microduck" it designed every part, checked nothing crashes into itself when it moves, did the motor maths so the thing can actually stand up, and handed me the parts list with prices and the step by step assembly not a render, a robot i can build this weekend every piece is ready to send straight to a 3D printer, and the motors and screws come with prices and where to buy them took me 45 minutes and 5 prompts to build (burnt all my limits) process below
Ronin23,292 views • 1 day ago
here are 5 robots you could build after reading this article: 1. the SO-101 arm, $122 in parts build it, calibrate every servo, teleoperate it, then print your own gripper fingers 2. a ROS 2 rover that maps a room and navigates it SLAM Toolbox and Nav2, and you can do the whole thing in simulation with no hardware at all 3. self-balancing robot it will not stand up until your filter and your loop timing are both correct, which is exactly the point 4. an ACT policy trained on your own demonstrations record 50 demos, train, deploy, measure the success rate, then record 50 more and beat it 5. line follower with PID your first closed loop, and filming it badly tuned next to properly tuned is the whole interview answer that's what will make you at least advanced junior at Robotics Engineering and will help you to get your first offer in robotics company
Ronin42,404 views • 4 days ago
5 startup ideas you can build and resell using only ElevenLabs Agents each one costs $0.08/min to run and replaces $2-5k/mo in human labor Let's break them down ↓ 1. AI Receptionist for Local Businesses dentists, salons, clinics, they all pay $2-3k/mo for someone to answer phones build a voice agent that: - answers calls 24/7 - books appointments - handles FAQs - speaks the client's language who ALREADY uses it: ~31% of local service businesses who STILL needs it: ~69% (your market) white-label it, charge $300-500/mo per client your cost per client: ~$30/mo in minutes 2. Multilingual Customer Support ElevenLabs agents speak 70+ languages natively e-commerce brands selling internationally need support in 5-10 languages minimum one agent replaces a 5-person multilingual team who ALREADY uses it: ~36% of e-commerce businesses who STILL needs it: ~64% and most of them are mid-market brands scaling globally sell 24/7 coverage, mark up the minutes, charge per-seat 3. AI Sales Qualifier (SDR Replacement) voice agent calls inbound leads, asks 5-10 qualifying questions, books meetings directly into the sales team's calendar startups pay $4-6k/mo per SDR you charge $1.5k/mo for an agent that works 24/7 and never misses a lead who ALREADY uses it: ~27% of mid-market teams who STILL needs it: ~73% and 22% already fully replaced human SDRs plug it into any CRM like HubSpot, Salesforce, Pipedrive 4. Restaurant Order-Taking Agent phone ordering for restaurants, pizzerias, takeout spots the agent takes the order, upsells sides and drinks, confirms, pushes to the POS who ALREADY uses it: ~34% of restaurants who STILL needs it: ~66% (expected to hit 50%+ in major cities this year) build one integration template → sell to 100+ restaurants at $200/mo each that's $20k/mo from one vertical 5. Real Estate Showing Scheduler agents answer property inquiry calls, give listing details, qualify buyers, and book viewings (all mid-call) realtors spend hours on phone scheduling who ALREADY uses it: ~18% use voice AI specifically who STILL needs it: ~82% while 82% of agents already use some form of AI, almost none have voice agents charge per listing or flat monthly integrates with their calendar + CRM -------- How to build any of these: - sign up for ElevenLabs (startups get $4k free credits) - pick your niche - build the agent with their no-code platform - connect it to GPT or Claude for the brain - plug in scheduling/CRM via API - white-label it under your brand you don't need to build AI, you need to sell AI to people who don't know it exists yet reply "ELEVEN" + RT and i'll send you a free guide so you can build this too
Ronin775,653 views • 4 months ago
![🚨 SOMEONE JUST KILLED THE COACHING INDUSTRY a developer spent 22,000 hours building a Personal AI Operating System on top of Claude Code now anyone with a terminal can install it for FREE it knows your goals, remembers every decision you've made, and prepares your morning briefing while you sleep [ the numbers are insane ]: - hours of dev work in it: 22,000 - sessions logged: 6,000 - time saved per day: 2-3 hours - GitHub stars: 12,100 - skills built in: 45 - workflows wired up: 171 - safety hooks: 37 - cost to install: $0 [ the science is wild too ]: no embeddings, no vector databases, no AI magic you can't read every memory, decision, and context lives in plain markdown files you read it with cat, search it with ripgrep, version it with git 4 memory types compound over time: - work memory (active projects, open decisions) - knowledge memory (domain expertise, research) - people memory (contacts, companies, relationships) - learning memory (patterns, mistakes, what works for YOU) every complex task routes through a 7-step cycle: OBSERVE → THINK → PLAN → BUILD → EXECUTE → VERIFY → LEARN privacy is enforced by CODE, not prompts a hook called ContainmentGuard physically blocks sensitive data from being written outside designated zones [ the grift opportunity is even wilder ]: freelancers are already charging $500-2,000 per personal AI setup for executives, founders, and busy operators one person + one weekend = a consulting business that didn't exist 6 months ago every AI productivity app you're paying $30/month for is replaceable by 4 hours of setup work and this one repo REPO: 100% OPEN SOURCE, FREE](https://image.24vids.com/tw-2053874586765152727/amplify_video_thumb/2053874150607892480/img/8Br269epANYy4SqZ.jpg)
🚨 SOMEONE JUST KILLED THE COACHING INDUSTRY a developer spent 22,000 hours building a Personal AI Operating System on top of Claude Code now anyone with a terminal can install it for FREE it knows your goals, remembers every decision you've made, and prepares your morning briefing while you sleep [ the numbers are insane ]: - hours of dev work in it: 22,000 - sessions logged: 6,000 - time saved per day: 2-3 hours - GitHub stars: 12,100 - skills built in: 45 - workflows wired up: 171 - safety hooks: 37 - cost to install: $0 [ the science is wild too ]: no embeddings, no vector databases, no AI magic you can't read every memory, decision, and context lives in plain markdown files you read it with cat, search it with ripgrep, version it with git 4 memory types compound over time: - work memory (active projects, open decisions) - knowledge memory (domain expertise, research) - people memory (contacts, companies, relationships) - learning memory (patterns, mistakes, what works for YOU) every complex task routes through a 7-step cycle: OBSERVE → THINK → PLAN → BUILD → EXECUTE → VERIFY → LEARN privacy is enforced by CODE, not prompts a hook called ContainmentGuard physically blocks sensitive data from being written outside designated zones [ the grift opportunity is even wilder ]: freelancers are already charging $500-2,000 per personal AI setup for executives, founders, and busy operators one person + one weekend = a consulting business that didn't exist 6 months ago every AI productivity app you're paying $30/month for is replaceable by 4 hours of setup work and this one repo REPO: 100% OPEN SOURCE, FREE
Ronin614,957 views • 3 months ago
somebody built a live feed of validated startup ideas you can copy here is the link: hundreds of early-stage AI products, most too new to be covered anywhere every tool on there is a founder who already tested a market and proved demand exists.. you're literally watching product-market fit happen in real time pick a direction that's already working, build your own take, skip the "will anyone pay for this" phase entirely while you don't have an access at least to $20M capital, no sense to be creative just take what works, add 1% unique proposition and cover the market (i've already spotted a few directions for my next products btw)
Ronin112,430 views • 26 days ago
Andrej Karpathy: "90% of Claude's mistakes come from missing context, not a weak model." 41% mistake rate without a CLAUDE.md. 11% with the 4-rule baseline. 3% with the 12-rule version below here are the 12 rules senior engineers settled on: 1. think before coding: state assumptions, don't guess. the model can't read your mind, stop hoping it will 2. simplicity first: minimum code, no speculative abstractions. the moment you let Claude add "for future flexibility," you've added 200 lines you'll delete next quarter 3. surgical changes: touch only what you must. don't let it improve adjacent code, that's how PRs blow up 4. goal-driven execution: define success criteria upfront, loop until verified. without them Claude either loops forever or stops too early 5. use the model only for judgment calls: classification, drafting, summarization, extraction. NOT routing, retries, status-code handling, deterministic transforms. if code can answer, code answers 6. token budgets are not advisory: per-task 4000, per-session 30000. by message 40 of a long debug, Claude is re-suggesting fixes you rejected at message 5 7. surface conflicts, don't average them: two patterns in the codebase? pick one. Claude blending them is how errors get swallowed twice 8. read before you write: read exports, callers, shared utilities. Claude will happily add a duplicate function next to an identical one it never read 9. tests verify intent, not just behavior: a test that can't fail when business logic changes is wrong. all 12 of Claude's tests can pass while the function returns a constant 10. checkpoint every significant step: Claude finished steps 5 and 6 on top of a broken state from step 4. nobody noticed for an hour 11. match the codebase conventions: class components? don't fork to hooks silently. testing patterns assumed componentDidMount, hooks broke them without surfacing 12. fail loud: "completed successfully" with 14% of records silently skipped is the worst class of bug. surface uncertainty, don't hide it what actually compounds instead of the next framework: - the CLAUDE.md file as institutional memory across sessions - eval-driven changes, not vibe-driven - checkpoints over speed - explicit conflicts over silent blending - discipline over framework, every time - one repo, one rules file, no exceptions be a few rules ahead of AI twitter before this becomes mass-opinion study this
Ronin451,445 views • 3 months ago
Andrej Karpathy: "90% of what AI twitter tells you to learn will be dead in 6 months" Here are 10 things senior AI engineers stopped wasting time on: 1. AutoGen / AG2: moved to community maintenance, releases stalled. dead for production 2. CrewAI: demos well, breaks in production. engineers building real systems already moved off it 3. Autonomous agent pitches: the AutoGPT / BabyAGI wave is dead in product form. the industry settled on supervised, bounded, evaluated agents 4. Agent app stores / marketplaces: promised since 2023, zero enterprise traction 5. SWE-bench leaderboard chasing: researchers proved nearly every public benchmark can be gamed without solving the underlying task 6. Microsoft Semantic Kernel: unless you're locked into Microsoft enterprise stack, it's not where the ecosystem is heading 7. DSPy: philosophical merit, niche audience. not a general agent framework 8. Horizontal "build any agent" platforms: Google Agentspace, AWS Bedrock Agents, Copilot Studio. confusing, slow-shipping, the math still favors building yourself 9. Per-seat SaaS pricing for agent products: market moved to outcome-based. per-seat is already dead 10. The framework that went viral on HN this week: wait 6 months. if it still matters, it'll be obvious what actually compounds instead: - context engineering - tool design - orchestrator-subagent pattern - eval discipline - the harness mindset (harness > model, always) - MCP as the protocol layer be few steps ahead than your competitors and outperform this market till it became mass-opinion study this.
Ronin417,200 views • 4 months ago
The Most Comprehensive Codex tutorial on the internet right now even OpenAI's own team said it's one of the best in under 2 hours you'll learn how to use Codex + GPT-5.5 (imo the strongest combo available right now) from skill and plugin setup, to automated workflow building, to multi-threading practical demos if you're looking for the best practices for vibe coding, this is the one to watch big shoutout and respect to Riley Brown
Ronin225,396 views • 4 months ago
Here's what the Founder of Claude Code does before he starts ANY project: 1. Plans first, codes never, he goes back and forth with Claude on the plan until it's perfect. No code gets written yet 2. Creates a CLAUDE.md file (a simple doc that Claude reads every session so it knows your project, your rules, your style) 3. Gives Claude a way to verify its own work - For backend: write and run tests - For UI: take screenshots, check in browser Claude should never finish a task without proving it works 4. Sets up project-level permission rules in settings.json instead of skipping permissions entirely. Shared with the whole team 5. Five more preparation steps in VIDEO BELOW 6. Only then switches to auto-accept mode and lets Claude build The part most people miss: he doesn't treat Claude as a magic box that gets things right first try he treats it like a junior dev that needs clear instructions, feedback loops, and guardrails at scale he runs multiple sessions in parallel, uses Opus with thinking enabled because it makes fewer mistakes even though it's slower and relies on background agents that push code for later review his setup is surprisingly simple. no crazy custom tools. just slash commands, subagents, and a clean CLAUDE.md the difference isn't the tool. it's how you set it up before you start ❤️ P.S. for sure 99.5% of readers will scroll down this tweet, but I send it to 0.5% who loves to learn how to improve your workflow daily and control 99.5% in 2 years Hard skills literally mean nothing in our world The most important skills which you can have are: building architectures and orchestrations If you master it and use a creative approach, CONGRATS
Ronin147,367 views • 4 months ago
i genuinely don't understand why nobody is building this.. last night, i built my growth operator system on an Obsidian vault with HUNDREDS of notes as markdown files: 1. every niche i've mapped, demand signals, supply gaps, which ones are heating up and which are already crowded 2. every creator the bot scouted, why their work is strong, what's broken in their distribution, who i passed on and why 3. every partner's offer, price, rev-share terms, and the objections that came up before signing 4. the content data, which topics pulled views vs which actually sold, hook by hook, with numbers next to each 5. the playbooks, scouting prompts, the content engine, whop CLI commands, outreach that doesn't sound like outreach 6. the rulings, every decision i never want to re-argue, one dated line each.. never below 25% rev-share, never partner without proof of skill 7. the trend radar, weekly scans of what's going viral in the niche right now, new formats and hooks that convert THIS week, not three months ago.. content stays on trend, conversion holds then whenever Grokbot starts working on a new partner, i point it at the vault it doesn't start from zero.. it starts with everything the last 10 partners taught it that's the difference between an agent doing tasks and an operator with MEMORY every partner makes the next one cheaper to run in the article below i break down the full growth operator system this vault plugs into:
Ronin16,032 views • 14 days ago
![This FREE guide shows you how to build a full automated TikTok content machine with Claude Opus 4.7 one setup. 30 slideshows a week. $0 cost [ by the end you'll have ]: - a library of proven viral hooks - a Pinterest-to-slides pipeline that runs on autopilot - a full week of content scheduled in 2 hours - a system you can reuse every single week [ the workflow ]: > steal hooks from viral TikToks > pull images from Pinterest for free > generate slides with a script (no Canva) > Postiz schedules the drafts for a full week IDEA: you can sell it as a service for $200-300/mo for each creator setup this system one time and onboard 10+ clients (you can do it easily, since it's needed to any creator) most creators still do this manually.. this automates the entire pipeline save it. set it up this weekend](https://image.24vids.com/tw-2044880255102443604/amplify_video_thumb/2044879799584260096/img/492KncEkQUZZsoIV.jpg)
This FREE guide shows you how to build a full automated TikTok content machine with Claude Opus 4.7 one setup. 30 slideshows a week. $0 cost [ by the end you'll have ]: - a library of proven viral hooks - a Pinterest-to-slides pipeline that runs on autopilot - a full week of content scheduled in 2 hours - a system you can reuse every single week [ the workflow ]: > steal hooks from viral TikToks > pull images from Pinterest for free > generate slides with a script (no Canva) > Postiz schedules the drafts for a full week IDEA: you can sell it as a service for $200-300/mo for each creator setup this system one time and onboard 10+ clients (you can do it easily, since it's needed to any creator) most creators still do this manually.. this automates the entire pipeline save it. set it up this weekend
Ronin102,935 views • 4 months ago
i merged all my AI agents into one brain Claude Code, Codex, Gemini: one team, one memory, zero re-briefing result: 8+ hours a week back, and every agent already knows my projects, my clients and how i think here's the setup with Jarvix: 1. dump your context once: projects, clients, writing style, how you make decisions.. it builds your personal model from your actual AI interactions 2. switch agents mid-task with zero re-briefing: Claude Code builds, Codex reviews, Gemini researches, all reading the same context 3. it carries what YOU choose across sessions.. never start a chat from zero again 4. when it's time to act, it pulls the right agents together as one team instead of five separate assistants the unlock isn't another agent. it's that your context finally travels agents without shared memory are five interns on their first day, forever. agents with it are a team that's worked with you for years not more agents. one shared context that's the right bet P.S. thanks to the Jarvix team for the early access
Ronin15,071 views • 25 days ago
This 1-hour Stanford lecture will teach you how to automate your entire daily work routine with AI better than 99% of "AI automation" content on YouTube No tools to sell you. No fluff. Just the raw science behind how AI agents think, plan, and execute tasks on their own Led by GitHub's Principal ML Researcher and recorded for FREE Most people spend 2-3 hours/day on repetitive work that AI can handle: emails, reports, scheduling, code reviews, data entry, status updates This lecture shows you exactly how the systems behind that automation actually work like reflection, planning, tool use, and decision-making Bookmark it. 1 hour now = hundreds of hours saved later
Ronin62,292 views • 4 months ago
![Here's how I'm running automated content engine in 2 files 1 markdown file = my wiki 1 html file = my dashboard that's the whole stack. [ the architecture, in plain words ]: LLM wiki = a single markdown file holding my audience DNA, 15 tracked creators, every viral topic from the last 30 days HTML artifact = a single page that reads that markdown file AND can trigger my agents the artifact and the agent talk to each other directly the wiki is the shared brain [ what I actually see when I open it at 9am ]: > 5 trending topics ranked by my audience-DNA fit > 3 KOL posts worth quoting today > last week's saved tweets (so I can ride waves that are still warm) > buttons: [draft tweet] [draft QT] [schedule] [log idea] 1. I click "draft tweet" on a topic 2. the artifact pings my agent 3. agent reads the wiki, drafts in MY voice, returns it to the artifact 4. I edit, schedule, done 15 minutes from morning coffee to 3 scheduled posts [ how to build the same in one evening ]: > step 1: dump your domain knowledge into ONE markdown file (audience profile, KOL list, content rules, voice guide, anything an agent would need to do YOUR job) > step 2: ask claude to build an html artifact that reads from that file ("here's my wiki, build me a dashboard with these views") > step 3: add buttons for the actions you do daily (draft, schedule, log, score, search — your workflow, not mine) > step 4: wire each button to call your agent via tool calls (so the artifact and the agent talk directly) the moment your artifact reads your wiki AND triggers your agents.. most SaaS tools you currently pay for quietly become unnecessary dashboards I used to pay $50/month for now sit in a single html file I can rebuild in 20 minutes every "I'll build a SaaS for this" idea you had last year is a 200-line file you write in an afternoon if you want to get the same content engine, just reply "CONTENT" and will send you in DMs later we're going from buying software to owning it.](https://image.24vids.com/tw-2053069887266771190/amplify_video_thumb/2053069374546710532/img/hxdHKnJf0ezvg9S_.jpg)
Here's how I'm running automated content engine in 2 files 1 markdown file = my wiki 1 html file = my dashboard that's the whole stack. [ the architecture, in plain words ]: LLM wiki = a single markdown file holding my audience DNA, 15 tracked creators, every viral topic from the last 30 days HTML artifact = a single page that reads that markdown file AND can trigger my agents the artifact and the agent talk to each other directly the wiki is the shared brain [ what I actually see when I open it at 9am ]: > 5 trending topics ranked by my audience-DNA fit > 3 KOL posts worth quoting today > last week's saved tweets (so I can ride waves that are still warm) > buttons: [draft tweet] [draft QT] [schedule] [log idea] 1. I click "draft tweet" on a topic 2. the artifact pings my agent 3. agent reads the wiki, drafts in MY voice, returns it to the artifact 4. I edit, schedule, done 15 minutes from morning coffee to 3 scheduled posts [ how to build the same in one evening ]: > step 1: dump your domain knowledge into ONE markdown file (audience profile, KOL list, content rules, voice guide, anything an agent would need to do YOUR job) > step 2: ask claude to build an html artifact that reads from that file ("here's my wiki, build me a dashboard with these views") > step 3: add buttons for the actions you do daily (draft, schedule, log, score, search — your workflow, not mine) > step 4: wire each button to call your agent via tool calls (so the artifact and the agent talk directly) the moment your artifact reads your wiki AND triggers your agents.. most SaaS tools you currently pay for quietly become unnecessary dashboards I used to pay $50/month for now sit in a single html file I can rebuild in 20 minutes every "I'll build a SaaS for this" idea you had last year is a 200-line file you write in an afternoon if you want to get the same content engine, just reply "CONTENT" and will send you in DMs later we're going from buying software to owning it.
Ronin50,315 views • 4 months ago
f*ck it. i'm LEAKING my entire Seedance 2.5 AI UGC system i cracked the formula for generations that look like real UGC filmed on a phone.. no camera, no actor, no set video below has been generated 100% by Seedance 2.5 + Claude... here's the FULL system: 1. open Seedance 2.5 and upload 3 references: Image your character, image2 the environment, David Carbon the style frame 2. assign every file a role in the prompt: "Image is the main character, image2 is the location, match David Carbon for the look".. never leave a file unassigned or the model guesses 3. write a timestamped storyboard instead of one messy paragraph: "0-8s: wide shot, she picks up the product. 9-16s: close-up, reaction. 17-30s: talks to camera".. the model follows it beat by beat like a shooting script 4. kill the AI look with one line: "retain real micro pores and skin texture, natural imperfections, no beauty retouching, no commercial polish" 5. add global rules at the end: "no subtitles, no fast cuts, same person in every frame" one generation. 30 seconds of footage. consistent character, zero drift now how to turn it into money, fast: > faceless youtube: stack scenes into 8-10 min docs. history, true crime, luxury pay $8-15 RPM, channels clear $5-20k/mo > UGC ads for brands: DTC brands pay creators $300-2,000 per video. yours cost minutes > sell the service: run this exact system for local businesses as a monthly retainer and the timing is unfair: Seedance 2.5 went live on Higgsfield today with 33 days of unlimited generations at zero credit cost on your plan a full month of unlimited practice and library-building before anyone else catches on most people will scroll past this don't be most people next i guess to make it automated P.S. thanks Higgsfield for sponsoring this post
Ronin14,935 views • 1 month ago
How to use 50+ API keys (models) for FREE on OpenClaw API??? - go to - login or register your account - click on "more models" - click on "use case" and select what you need it for - choose the model and open it - click on "view code" → "Generate API key" many models don't allow direct deploy, so use the "view code" button to generate API access basically Nvidia NIM gives you the ability to test almost any model from their list for FREE some of them are not worse than GPT 5.2 or Claude Opus 4.6, some might even perform better depending on the task how to understand if a model is efficient and compare it with others??? - go to - type the model name in search - click on "benchmarks" - you’ll see performance tests and rankings this way you can easily compare free models with paid ones of course there are RPM limits, on many models it’s around ~40 requests per minute each model is different, after generating the API key, RPM limits are shown in the top-right corner nothing stops you from using them, many work perfectly fine, super solid option for first tests and for learning OpenClaw or any other system where you need an AI API model
Ronin58,974 views • 6 months ago
Jane Street pays $750k/year for quants who can run neural networks across thousands of market signals This 1-hour Cornell lecture by Marcos Lopez de Prado gives you the same framework those quants get paid $60k/month for Bookmark & watch today. Then read the article below to repeat
Ronin35,193 views • 3 months ago
Zach Yadegari (Cal AI founder) built a $1.12M/month app at 17 with no paid ads and no VC 80 minutes on how he went from coding in his bedroom to 8M+ downloads while still in high school worth more than any $2k "how to build a startup" course here's what he covers: > find a simple problem people already pay to solve badly > build the MVP fast, make it stupid simple to use > skip ads entirely, go influencer-first from day one > scale with creators, not capital the founders who win right now aren't raising rounds.. they're shipping apps from their laptop and letting creators do the marketing that's exactly why I put together 20 startup ideas you can build the same way the full list is in the tweet below ↓
Ronin25,192 views • 3 months ago