Darius Reginis

1K posts

Darius Reginis

Darius Reginis

@drskull

Games developer, frontend lover, #bitcoin hodler

Lithuania Katılım Şubat 2009
1K Takip Edilen140 Takipçiler
Vishal
Vishal@VishalMalvi_·
Big salute to this brave Indian petrol pump worker , handled the bike fire calmly and saved the situation
English
262
699
15.8K
3.3M
осфрид 🎲
осфрид 🎲@FreddSnow·
как здорово, что в делегацию пасе прошел марк фейгин, который (по его же словам) в качестве "авантюры" ездил совершать военные преступления в боснийской войне добровольцем на стороне сербов под командованием ратко младича и участвовал в осаде сараево. и всем на это насрать 👍
Русский
64
35
813
68.9K
Darius Reginis retweetledi
Guillermo Rauch
Guillermo Rauch@rauchg·
This is fun. Left the autonomous chess match (v0-chess-match.vercel.app) running overnight between @xai Grok 4 and @openai GPT 5.2. Grok's 𝚐𝚛𝚘𝚔-𝟺-𝚏𝚊𝚜𝚝-𝚛𝚎𝚊𝚜𝚘𝚗𝚒𝚗𝚐 won 19 of the last 20 matches 😅 What models should battle next?
English
445
535
5K
4.4M
Leonid Volkov
Leonid Volkov@leonidvolkov·
Сегодня в центре интернет-бури дня — мое частное сообщение бывшей сотруднице, которое она решила опубликовать. Даже специально фейсбук для этого завела. Хочу сказать по этому поводу несколько слов. Я гражданин России и российский политик. Много лет я, отстаивая интересы россиян, борюсь с преступным путинским режимом, узурпировавшим власть в нашей стране. Разумеется, я ни на секунду не забываю о том, что путинский режим начал четыре года назад преступную полномасштабную войну против Украины. Путинская агрессия — безусловное зло, которое должно быть побеждено. Мои усилия и усилия моих коллег в этом направлении сосредоточены на изменении настроений в России, на лишении Путина, его режима и его войны любой поддержки. В то же время я продолжаю считать — эту позицию я публично высказывал, и буду дальше отстаивать — бессмысленной и контрпродуктивной кампанию по раздуванию ненависти ко всем россиянам, по навязыванию чувства коллективной вины по цвету паспорта. Эта кампания — большой подарок Путину и его пропаганде, она помогает ему консолидировать вокруг себя остатки поддержки. Я много раз говорил и буду говорить о том, что ту энергию, которая была направлена на травлю российской политической оппозиции, куда полезнее было бы направить на борьбу с Путиным. И, конечно, к организаторам и участникам этой травли я отношусь очень плохо. Что же касается моего личного сообщения, которое сегодня обсуждают, то оно, конечно, эмоциональное и резкое, как это, увы, нередко бывает с личными сообщениями. Оно было адресовано бывшей сотруднице, которая занимается постоянным распространением гадостей и лжи про меня и про ФБК, присоединившись к кампании против российской оппозиции. Мне не следовало писать это сообщение, и, конечно, надо было лучше контролировать свои эмоции. P.S.: И да, я плохо отношусь к неонацистам. Если для кого-то это сюрприз, ну извините. Даже тактический союз с ними мне кажется большой ошибкой.
Русский
540
79
1.4K
199.1K
Alton Syn
Alton Syn@WorkflowWhisper·
Claude and Cursor can now build n8n workflows directly in your instance. No JSON. No copy-paste. No 3-hour debugging sessions. And it's about to make $15K consultants absolutely obsolete. Most AI tools generate broken templates and say "good luck figuring it out." Synta's MCP works completely different: 1. IT INTERVIEWS YOU BEFORE BUILDING "What's your preferred database—Airtable, Google Sheets, Notion?" "Where's your data coming from—Apollo, your CRM, web scraping?" "How should errors be handled?" It beefs up your prompt before writing a single node. While consultants charge $200/hour for "discovery calls"... This does it in 30 seconds. For free. 2. IT KNOWS N8N BETTER THAN YOUR CONSULTANT Scrapes the latest documentation in real time. Knows what nodes exist TODAY. Not what existed when Claude was trained in 2023. Your $12K "expert" is still googling deprecated node parameters. This thing has the entire n8n docs embedded. Updated. Current. 3. IT BUILDS DIRECTLY IN YOUR INSTANCE No exporting. No importing. No praying the nodes connect. No "let me check if this JSON is valid." Describe what you want → watch it appear in your n8n → ready to run. I told it: "Build a competitor monitoring system that scrapes pricing, analyzes with AI, and alerts me when I'm being undercut" 47 seconds later: 3 workflows. Sitting in my n8n. Connected. Running. The consultant quote for this exact system? $14,000. My cost? $0. This is what my agency runs on daily. 10 people operating like 50. Same-day delivery on projects that used to take weeks. While other agencies are posting job listings for "Senior n8n Developer - $120K"... We're shipping faster with half the team. Comment "MCP" and I'll send you: → How to connect Synta MCP to Claude or Cursor → The prompting structure that gets clean builds every time → Link to check if you got early access (~50 accounts only) The automation consulting industry just lost their biggest advantage. And I'm handing it to you for free.
English
398
50
585
55.3K
Nozz
Nozz@NoahEpstein_·
the n8n tool nobody's talking about just mass-deployed 14 workflows to my instance in 8 minutes while everyone's still copy-pasting JSON and debugging broken nodes here's what's different about this one: every AI tool that claims to "build n8n workflows" does the same thing: → generates JSON based on outdated training data → uses nodes that don't exist anymore → forgets how authentication actually works in 2025 → hands you a file and says "good luck" i've tested all of them: - claude desktop with mcp - chatgpt with custom instructions - gemini with n8n docs pasted in - cursor building directly in the repo they all guess at n8n structure this one actually knows it here's the difference: it scrapes n8n's live documentation in real-time not documentation from when the model was trained documentation from TODAY when n8n updates a node → it knows immediately when authentication flows change → it adapts when new integrations drop → it's already using them built a full lead gen workflow: webhook → enrichment → crm sync → slack notification → email sequence 4 minutes. 14 nodes. zero errors on first run. same workflow with claude took me 45 minutes and 3 rebuilds then i stress tested it: "build me a client onboarding system with stripe, notion, gmail, and slack" watched it: → plan the entire workflow architecture → build each node with correct field mappings → deploy directly to my n8n instance → run it → hit an auth error → debug itself → fix the oauth scope → run again → work perfectly 8 minutes. live in my instance. ready for clients. the thing that breaks my brain: it doesn't just generate a workflow it deploys it, runs it, watches it fail, fixes it, and runs it again until it works most AI tools give you a starting point this one gives you a finished product here's the reality check: agencies charging $8-15K for n8n builds are using tools that guess while a tool exists that actually knows the gap between "AI-assisted" and "AI-built" just got real comment "BUILD" and i'll send you the tool + the exact prompts i use to get production-ready workflows in under 10 minutes
English
504
41
433
42.6K
Lian Lim | Dashboard & AI Automation Expert
i built 47 n8n workflows in 3 months everyone asks: "how do you build so fast?" ONE prompt i created a mega prompt that tells any LLM how to architect n8n workflows the process: → copy prompt → describe what you want → get production-ready spec no more: - spending weeks figuring out structure - googling "where does error handling go" - wondering if you need code or built-in functions - rebuilding because you missed something the AI generates: → complete workflow architecture → node-by-node specs → error handling built in → API integration details → testing cases → deployment checklist works for ANY automation: like order processing, CRM sync, email sequences, document workflows, approval chains, calculations same prompt structure every time reality check: most people will save this and never use it but if you actually use it? 50+ hours saved this year = $5K-$25K in time back the prompt is free. inside: → complete mega prompt (copy-paste) → 3 example workflows → works with ChatGPT/Claude/any LLM → common mistakes to skip → advanced customization → 10-step checklist → ready-to-use templates WANT the prompt playbook on how to build any n8n workflow with AI? Like + Comment “FREE” and i’ll dm it to you. some people are building agencies around this.
Lian Lim | Dashboard & AI Automation Expert tweet media
English
924
215
1.8K
163.2K
Darius Reginis retweetledi
Kit Langton
Kit Langton@kitlangton·
Visual Types, a humble collection of semi-interactive TypeScript lessons, is ready to be clicked upon. types.kitlangton.com ← 👀
English
53
127
1.1K
154.3K
Yulia Latynina
Yulia Latynina@YLatynina·
The problem is not that Ukraine has been part of Russia for centuries. The problem is that Ukraine in its 1991 borders has been created by the expansion of Russia, the same way Texas is created by the expansion of US. Suppose for some reason back in 1991 the USSR won, parts of US seceded, and Texas became independent. There’s no inherent harm in this. But then suppose you hear that there’s a “decolonization” of Texas going on, that schools in English are prohibited, that you should speak only Algonquin, that English is the language of colonizers and that any Texan who made a career in US proper is a traitor to the nation. All the activists that are demanding this are, for some strange coincidence, financed by Komintern NGOs, and Texas itself is to join the Warsaw pact. And then, when the dismembered US goes to war with Texas, it hears that it is an unjust aggression, that all US citizens are complicit in it, and that whatever the sovereign state of Texas was doing on its territory, is none of the US business. That’s soft power for you. And yes, rest assured: in the above example the Komintern NGOs won’t have the best interests of Texans at heart. They’ll have just one goal: to make US and Texas destroy each other. And they will reach this goal with minimum expenditure and zero risk.
English
482
52
431
234.6K
Darius Reginis retweetledi
Lu Wilson
Lu Wilson@TodePond·
can't stop playing with this
English
52
62
1.3K
119.6K
Darius Reginis retweetledi
Yohei Nishitsuji
Yohei Nishitsuji@YoheiNishitsuji·
for(float i,g,e,s;++i<139.;){vec3 p=vec3((FC.xy-.5*r)/r*.1+vec2(0,1.18),g-1.4);p.zx*=rotate2D(t*.3-9.);s=1.;for(int i;i++<26;p=vec3(.33,4,-.64)-abs(abs(p)*e-vec3(2.7,3.95+e*.15,3.5-g)))s*=e=5./dot(p,p*.5);g+=p.y/s;s=log(s)*g+g;o.rgb+=.01-hsv(e/i-.38,p.y*.4,s/1e3);}s;#つぶやきGLSL
15
85
757
24.4K
Darius Reginis retweetledi
Roman
Roman@romanroan·
@prof_preobr Reminds me of this…
Roman tweet media
English
4
2
62
2.9K