cap't (@JariusOS) header
cap't (@JariusOS) profile picture

Building @warplette & @jujuisr for endangered species & African Commodities.

mama africa
Joined November 2021
Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)Photo by cap't (@JariusOS)
Our half-million parameter model just cleared all five waves in The Lost Storm. It still ignores gold, but this is a big step for minions run by the model. The Lost Storm is free in your browser. theloststorm.com
1
1
2
46
cap't retweeted
DNA Markets will come to X.
X is the best source for financial news -- and hundreds of billions of dollars are deployed based on things people read here. We are building Smart Cashtags that allow you to specify the exact asset (or smart contract) when posting a ticker. From Timeline, users will be able to tap them to see its real-time price along with all mentions of that asset. We're aiming to collect feedback as we iterate toward a public release next month.
1
2
43
10 Claude Code tricks most people never use. Number 3 alone will save you hours. Save this for your next session.
6
2
61
40,175
cap't retweeted
𝗖𝗼𝗱𝗲𝘅 𝘁𝗶𝗽: if you still have half your quota left when it resets, you’re throwing away days of Astra and Sol. Use `/goal` to squeeze every last bit out of it and clean out what months of vibe coding piled up: dead code nothing calls, the same logic written several times over, dependencies you don’t need, old sessions and worktrees hogging your disk, or even scheduled tasks that have been running for nothing. Here are 5 prompts and goals you can copy-paste directly. ↓
62
91
8
1,669
263,462
running a company nowadays feels kinda dystopian my team is 50+ people in 12 countries, but i talk to my agents more than my human teammates - voice note to grokbot - mattpocock/skills /grill-me so the agent interrogates the idea before anything gets built - spec drops into linear through the linear mcp - exa + firecrawl agent reads every competitor site, pricing page and changelog - ideabrowser.com to find ideas and trends - apify pulls g2, reddit and app store reviews - prompt: "find the 10 complaints that show up most and quote the users" - context7 so the agent reads the latest docs before it touches any api the design - claude code + paper mcp - 4 versions of every screen before a single line of code - pick pieces from 2, it merges them. diverge then converge - frontend-design skill from anthropic/skills + our own design system skill - cleanshot screenshots of any ui i like go straight in with "make it feel like this" the build - pingdotgg/t3code as the control surface for every agent, from my phone too - stablyai/orca runs a fleet of agents in parallel, each in its own git worktree - graphify maps the whole codebase into a knowledge graph so agents navigate instead of guess - headroom compresses tool outputs before they hit the model, so tokens go 20-95% further - lsp plugins so the agent sees types and errors like an ide does - security guidance hook flags risky edits before they land - supabase mcp for the database, stripe for billing - github mcp opens the pr, vercel spins up a preview - loops run overnight, i check what shipped over coffee the user testing - lightpanda as the headless browser, way lighter than chromium so you can run hundreds at once - agents click every flow on the preview like a first-time user - one persona is impatient, one is confused, one is on a phone - playwright + chrome devtools mcp for regression tests the bugs - sentry mcp catches the error - agent reads the stack trace and opens a fix pr in under 10 minutes - human just approves the data - posthog mcp cron at 7am on yesterday's funnel - any step that drops more than 10%, the agent watches 20 session replays and tells me why - stripe mcp pulls every cancellation reason into one doc every monday the agents' tools - orgo so every agent gets its own computer - agentmail so every agent gets its own inbox - agentphone so i can call them from the car - composio connect-apps for every other connector - akitaonrails/ai-memory, a git-backed markdown wiki so they remember what we decided last week the skills - wshobson/agents so skills work across claude code, codex and opencode - my writing voice - our design system - how we write specs - skill creator so the agents write new skills themselves some helpful tricks - one agent per platform. the x agent never touches email - cheap model for triage, opus for anything that ships - every output gets scored 1-10 by an llm judge. under 8 goes back - every agent writes a 5-line log at night so i can read the whole company in 2 minutes - every agent mistake becomes a new skill, so it only happens once the team - humans get version 5, not my voice note which they are happy about lol - their job is creative feedback, relationships and telling the agents what they got wrong a human pm reviews the funnel once a week. this reviews it every morning running a company is now voice notes and inference
577
333
153
5,854
745,839
No revenue, users, or even a product... These 10 funds will still write you a cheque (yes, you read it right): 1. Afore Capital (@AforeVC) - a $500M fund that only does pre-seed. Backs you before traction, even before a company 2. Kima Ventures (@2lr) - Xavier Niel's fund. 2 startups a week, ~100 new deals a year 3. Pareto Holdings (@edwardlando) - invests only its own money. Edward has been early in 35+ unicorns 4. Precursor Ventures (@chudson) - often your first institutional cheque, before the product even exists 5. Village Global (@villageglobal) - pre-seed network backed by Bezos, Gates and Zuckerberg 6. Pear VC (@pearvc) - pre-seed + seed, runs PearX, cheques from $250K 7. Boldstart (@Boldstartvc) - "the inception fund". First cheque for technical founders in dev tools, security + AI infra 8. Kindred Ventures (@stevejang) - Steve was early in Uber, Coinbase and Perplexity 9. Unshackled Ventures (@UnshackledVC) - pre-seed for immigrant founders, with visa support 10. Inflection Point Ventures (@ipventures_in) - 21,000+ angels, 250+ startups funded 🇮🇳 At pre-seed, nobody buys your numbers; they bet on you. Follow @arorabhavyam for more content around founder programs, investors and AI! 🫡
30
63
7
632
85,526
cap't retweeted
At @ycombinator, we're excited to fund more crypto startups solving financial needs including AI capital raising, stablecoins, trading, and agents. Here's our latest crypto request for startups. Our next application deadline is Monday.
380
284
290
3,124
726,674
5 years ago, this young chap explains why six months is enough to genuinely change your entire life. I wish I found this genius clip years ago.
73
3,280
37
19,288
510,579
cap't retweeted
🚀 100 Places to Launch Your Startup 1. producthunt.com 2. reddit.com 3. news.ycombinator.com 4. github.com 5. indiehackers.com 6. crunchbase.com 7. wellfound.com 8. g2.com 9. capterra.com 10. alternativeto.net 11. sourceforge.net 12. appsumo.com 13. getapp.com 14. saashub.com 15. trustradius.com 16. betalist.com 17. peerlist.io 18. f6s.com 19. slashdot.org 20. stackshare.io 21. theresanaiforthat.com 22. toolify.ai 23. futurepedia.io 24. futuretools.io 25. devhunt.org 26. uneed.best 27. hashnode.com 28. goodfirms.co 29. softwareadvice.com 30. softwaresuggest.com 31. crozdesk.com 32. startupranking.com 33. startupstash.com 34. feedmyapp.com 35. launchingnext.com 36. microlaunch.net 37. fazier.com 38. dang.ai 39. libhunt.com 40. sideprojectors.com 41. startupbase.io 42. peerpush.net 43. nicklaunches.com 44. tools.launchllama.co 45. startupfa.me 46. startupbuffer.com 47. tekpon.com 48. serchen.com 49. findly.tools 50. 10words.io 51. Aura++ 52. IndieDev Tools 53. NoCodeList 54. 500 Tools 55. OpenHunts 56. SaaS City 57. LaunchIgniter 58. SideHunt 59. PitchWall 60. FoundrList 61. Awesome Indie 62. OpenAlternative 63. BetterLaunch 64. Altern AI 65. Startup Inspire 66. Stacker News 67. AI Directories 68. Uno Directory 69. Open Launch 70. What Launched Today 71. Toolfolio 72. SaaS Browser 73. SaaS Genius 74. Rank Anything 75. Slocco 76. AI Agent Store 77. Startups FYI 78. Twelve Tools 79. Crazy About AI 80. EarlyHunt 81. TinyLaunch 82. Tiny Startups 83. Manta Directory 84. ConfettiSaaS 85. DokeyAI 86. IndieHunt 87. AcidTools 88. Lapa Ninja 89. MakerHunt 90. ToolFame 91. ShowMeBest AI 92. Good AI Tools 93. ScrollLaunch 94. Hada News 95. That AI Collection 96. Rank in Public 97. Firsto 98. TrustMRR 99. API List 100. Shipybara You don’t need a massive marketing budget to get your first customers. You need distribution. My strategy: → Start with 10 relevant platforms. → Prioritize free submissions. → Write unique descriptions. → Track referral traffic and signups. → Double down on what converts. Some platforms require payment or approval, so check their submission rules. Don’t just build your startup. Get it in front of people. Bookmark this list for your next launch. What website did I miss?
75
65
8
398
31,053
Meta Muse will probably create 1000+ millionaires This is the most slept on arbitrage I’ve seen in years and it’s not even close. Here’s how i’d build a 1 person business powered by AI/Muse with with $0 and 0 social following
492
768
92
10,957
1,459,105
cap't retweeted
Most people think building a personal brand takes 6+ months. It doesn't, and I proved it by building one in 9 days. In this video, I break down how I built a reputable personal brand in literally one week. (copy this to grow):
14
33
1
390
32,983
cap't retweeted
grok bot changes how i build companies going forward here are seven use cases you won't believe are possible
11
4
1
158
47,223
As CEO of Product Hunt, I see hundreds of launches a week. One trait separates winners from losers. Every few months, new models come out. Some products become better, effortlessly. Some don't improve, or become irrelevant. This property needs a name. I'm calling it Buoyancy. Buoyant startups rise with model improvements the way a boat rises with the tide. Anti-buoyant startups come in two flavors: over-scaffolded (which hinders the new, smarter models) and too thin - which makes them redundant when models improve. To be buoyant, you need a "hull" - usually proprietary data, a vertical-specific harness, network effects, or distribution advantage. This hull keeps you buoyant as models improve. Hulls are the new moats. My number one advice to startups in 2026: Be buoyant. Find your hull. (Full essay on my blog)
113
129
48
1,642
106,718
If you’re a first generation achiever or you’re trying to be one in your family, this one is for you!
196
2,803
212
11,760
776,233
cap't retweeted
看了一期 Starter Story,大意是讲如何 copy 人家的 idea 的,嘉宾 Samuel 以前是配镜师,跟着 YouTube 自学写代码,现在 3 个 SaaS 月入 3.5 万美元。 他的方法说穿了就一句:只做已经有人做成了的产品。 他挑点子有四个条件: 1. 自己会用 2. 已经被验证能赚钱,比如推特上有人晒 MRR、Stripe 截图 3. 对方没在营销上砸大钱,说明需求是真的 4. 产品够简单,一个人维护得了 最有意思的是他验证的方法:用 Ahrefs 查对方的流量从哪来。 主要靠广告的最好抄,你投一样的广告,一周就能有流量。 广告和 SEO 都有的,说明需求强。 主要靠 SEO 的最难,要熬时间。 他的 AI 视频工具就是这么来的:看到有人晒做 faceless 视频工具的收入,一查流量全来自 Facebook 广告,就照着做了一个,投广告,后来做到月入 2 万美元。 反过来,他第一个产品没用这套方法,做邮箱查找工具,正面撞上 Apollo 这种大公司,维护得特别痛苦。 当然,他复制的是被验证过的需求,不是别人的代码和设计。 他说最让他意外的一点是:根本不需要创新,简单无聊的工具最赚钱。 我自己的感受是,很多人做独立开发,总想找一个没人做过的点子。但对一个人来说,做一个已经被验证过能赚钱的东西,比起自己凭感觉写了个 app 然后却不知道卖给谁以及如何卖更实在。你们觉得呢?
40
155
14
951
71,579
Yapay zeka ile video üretiyorsanız ya da üretmeyi planlıyorsanız bu websitesine bir uğrayın. Çünkü iyi bir YZ videosu üretmenin yolu yalnızca oradan buradan iyi promptlar toplamaktan değil, sinemanın dilinden anlamaktan geçiyor. “Cinematic Techniques” adlı bu websitesinde film ve video yapımında kullanılan tam 424 farklı teknik bir araya getirilmiş. Sadece kamera hareketleri başlığı altında bile 86 farklı teknik var: Dolly In / Out, Tracking Shot, FPV Drone, 360-Degree Orbit, Dolly Zoom, Handheld... Ama asıl değerli tarafı, bu websitesinin sadece bir sinema terimleri sözlüğü olmaması. YZ video üreten bir servis içinde yer alan bu sayfalarda, kompozisyon, kamera açıları, aydınlatma, lensler, renk tonları, geçişler ve Film Noir gibi türsel anlatım biçimleri de sistematik biçimde ele alınıyor. Üstelik her tekniğin; → Nasıl kullanıldığı → Görüntü üzerindeki etkisi → Video örnekleri → Referans görüntüler → AI video üretiminde kullanılabilecek prompt’lar ile birlikte sunuluyor. Bence asıl potansiyel buradaki yüzlerce prompt’ı kopyalamaktan çok, bu kaynağı YZ destekli bir video üretim sistemine dönüştürmekte. Örneğin: “Endişe verici ve tekinsiz bir sahne oluşturmak istiyorum.” ↓ Uygun çekim tekniğini belirle. ↓ Camera Angle × Lighting × Lens × Color kombinasyonunu oluştur. ↓ Sahnenin görsel dilini tanımla. ↓ Kullanacağınız AI video aracına uygun prompt’ı üret. Hatta bu mantıkla bir Skill oluşturabilir veya sitenin içeriğini Claude Opus 5.5’e vererek onu bir görüntü yönetmeni gibi kullanabilirsiniz. Böyle baktığımızda YZ video üretiminde asıl beceri giderek değişiyor: Prompt ezberlemek değil; ne istediğinizi görsel bir dille tarif edebilmek. Çünkü YZ araçları geliştikçe prompt yazmanın teknik kısmı giderek kolaylaşacak. Ama hangi kamera hareketinin hangi duyguyu yarattığını, ışığın anlatıyı nasıl değiştirdiğini, lens seçiminin izleyiciyle karakter arasındaki mesafeyi nasıl etkilediğini bilmek, insana ait bir yaratıcı karar olmaya devam edecek. YZ video çağında öğrenmemiz gereken şey daha fazla prompt değil, daha sinema dili. #AI #AIVideo #AIMovie
13
197
4
1,023
57,979
cap't retweeted
the people that got cal.com to #1 in chatgpt made a video explaining how to win AEO heres everything they shared: >AI search isnt a black box. the models rely on consensus from the open internet and u can literally view the sources chatgpt uses to form its answers >"#1" doesnt mean winning one query. it means consistently showing up in the top 3 across dozens of different prompts >they wrote 47+ comparison blog posts like "cal vs calendly" and "cal vs savvycal" targeting ppl who are ready to buy, not ppl who are just browsing >they stopped writing generic informational content that AI already answers on its own. every page became product-led and problem-solving >1 in 3 cal.com sales meetings now comes from AI visibility. they track revenue changes not rankings >youtube has a massive correlation with AI search visibility. video content is harder to fake than text and AI models trust it more. youtube is the new reddit for AEO >they deleted thousands of thin and underperforming pages from their site. less content, better content, more traffic. sounds backwards but it works >they use a concept called parasite SEO -> leveraging high authority platforms like youtube and linkedin to rank content faster than they could on their own domain >mass producing thousands of AI generated articles backfires. it creates keyword cannibalization, triggers penalties, and signals low value to search engines >AI search is winner take all. if ur not in the top 3 recommendations u basically dont exist to the user. theres no page 2 >if ur starting from zero with no budget: skip the fancy CMS. make one high quality video comparing ur product vs a competitor and make sure ur technical SEO foundation is solid. u can start with $100 this is the most transparent breakdown ive seen from a team that actually did it. not theory. results.
29
14
2
510
52,916
The fastest way to get ahead is to get obsessed for 1 year. Pick a skill people pay for. Get extremely good at it. Build proof that you're good at it. Create content about it and put your proof online. Iterate until you get it right. That's how you can change your life in 1 year.
41
446
12
3,193
47,804
I woke up to 1,000+ followers today 🥹 It took me 8 months and around 3,800 posts. Here's what worked for me and what didn't. (Your mileage may vary, obviously) Some numbers first: → Started in February 26 → ~3,800 posts, so roughly 16 a day → Last week I sent 170 posts and 140 of them were replies → July: +11 followers. August and September: +360 → ~400k impressions in the last 90 days For months it felt like shouting into the void, and then somewhere around August it tipped So here's what didn't work: 👎 → Chasing big accounts Engage with your heroes, reply and repost them, that's all fine. Quoting a big account can even give you a real boost. I quoted @marclou post about his China trip and added my own experience, and it brought me ~100 followers and 100k+ impressions. Marc didn't follow me back and that's fine, I never expected him to. As a small account I just wouldn't build my strategy around getting noticed by them. They play Champions League, we play Sunday League - for now ;-) → Talking too much about myself tbh think my story is kinda interesting. I changed direction a bunch of times in my 20s, then at 30 I moved to Bangkok and started a business knowing nothing about business. I posted all of it with a photo of me leaving Germany, spent about an hour polishing it, and it flopped hard. People care, just a lot less than you'd hope when a post is all about you. → Lazy AI Don't be lazy and let Claude write your posts. People here sniff out AI posts right away and hate it. If you want to grow, you have to be real. You can use AI for that but set up a system with md files that know your tone and voice → Posts that feel scheduled I tried these at the beginning, those polished calendar wisdom, a hot take, a statement etc. And most of the time I asked AI to come up with these and that's why it didnt work. They might get some reach but they show nothing about who you are. My best posts were the ones I typed spontaneously and never planned What worked 🚀 → Being myself (by far the most important) Show your ups, your downs, your doubts and your confidence. Have interesting takes on things. If people can't figure out who you are, they have no reason to stick around. → Getting verified A blue checkmark is your signal to the community that you’re serious because the most toxic, destructive and stupid replies come from people without. No checkmark, no growth → Building my own tribe I don't bother if someone has 50 followers or 5,000. What matters is whether they feel like my crowd: they engage, they share their views, and I share mine on their posts. I'm pretty sure X rewards that kind of back and forth because it shows you're actually socializing here. (Which explains the 140 replies.) The best people I got to know here have an account my size and we're growing together → Curating my feed I told the algorithm what i want to read. Whenever theres a post I don't care about, I mark it as "not interested". By now my feed has no politics, no bait and no divisive stuff, just tech, AI and builders. It's so positive that I love coming back every day to engage, and it keeps me engaging with the right things → Riding a trend When Composer 2.5 came out I posted about how much faster it was than Opus and GPT. It got 1k+ likes, hundreds of thousands of impressions, and even @elonmusk liked it. If you have something real to say about a trending topic, say it fast → Sharing wins I once posted that I did $35k in sales in one week and it brought in a lot of attention and followers. I'm still torn on this one because it feels boasty to me, so I've been careful with it. But it works → Milestone posts My posts at 300, 500 and 900 followers all did well. People here are way more supportive than I expected and they really cheer for you. → "40 yo founder from NZ, looking to connect" kind of posts They work. I've never done them tho because the intention is so obvious: I want followers. And when someone comments "let's connect" and nothing else I mostly ignore it, because I learn nothing about them. I'm here to hang out with real people. And I think this is what X is all about If you're stuck under 500 and it feels like nobody's listening: it gets a lot more fun once you're past ~900 (at least in my case). Just be authentic, and genuinely engage with others. What worked for you so far? and what didnt?
107
6
8
194
8,342