duc le

@RaulJuncoV hi Paul, just wonder which tool do you use for the diagram?

Push-based systems come up in 90% of system design interviews. Here's the exercise you should be able to solve: Design a notification system for 100M users. Some have 50 followers. Some have 10M. The instinct is to hold a WebSocket connection open to every active user and push updates as they arrive. Clean mental model. It collapses the moment a celebrity posts. When someone with 10M followers posts, you push to 10M open connections simultaneously. Your message broker saturates. Your WebSocket servers fall over. The system fails at the exact moment it needs to work. That's the fan-out problem. And it kills more interview answers than any other mistake. The production answer: push and pull aren't binary. You pick based on follower count. Users with fewer than 1,000 followers get push fan-out. Each follower gets notified immediately. Users with millions of followers get pull fan-out. Their feed assembles on read. Nobody gets a push. Followers see the post when they open the app. Twitter built exactly this: push-on-write for small accounts, pull-on-read for large ones. But fan-out is only half the problem. Push means stateful connections. Your servers now need to know which connection lives on which machine. You can't route blindly. Most teams reach for Redis pub/sub here; the WebSocket server subscribes, the backend publishes, the message finds the right node. Add a 3-second network drop and you have another layer: what did the client miss? Now you need sequence IDs, a message buffer, and reconnect logic that replays missed events. "Push-based" became push with a pull fallback, a message broker, sticky routing, and a replay buffer. Most engineers stop at the first diagram. The ones who get the offer keep pulling the thread until the system breaks.

@andrewbrown I would love to join

Anyone want to test our Claude Code Essentials course in the ExamPro Platform in Early Access? Like this tweet, so I can pick a couple of folks (make sure your DMs are open). This is just to help us spot any obvious gaffs in the materials.

@jakezward I registered and please send report for my company

I just built SEO strategies for beehiiv and Kleo. A ~$300M newsletter platform and my own 7-figure SaaS we launched 43 days ago. Completely different products, audiences, and business stages. But I'm using the same playbook. SaaS SEO in 2026 isn't what it was 2-3 years ago. The old playbook: - Target high volume keywords - Build comparison pages - Rank in Google and get clicks But in 2026: 1. Search is as fragmented as ever across Google, LLMs, socials, and other platforms. 2. Even if you rank, SERPs are busy with AI Overviews and ads making position #1 less valuable. 3. Zero-click searches mean AI is answering queries before any clicks to your website. 4. AI has commoditised average content and the quality bar is much higher. I'm breaking down how I built the strategies for beehiiv and Kleo in a live session tomorrow at 12PM EST. You'll see: - What 20% of SEO to focus on in 2026 - How to rank across Google AND AI platforms - The exact playbook for turning SEO into revenue Register (for free) in the replies. PS. The next 20 SaaS companies that register and reply below get a free AI visibility report from my team (normally only for our clients like beehiiv).

@alexhalliday AIROPS

@alexgroberman SEO Stuff Cheat Codes

A recent study analyzed 75,000 brands to see what actually determines visibility in AI Overviews. In the process, they exposed some key ways SEO Stuff (seo-stuff.com) has been getting customers traffic + sales for months now. Now that none of it is a secret anymore, let's talk about it. But before we get into it... If you want some cheat codes that will have you showing up in AI Overviews within the next 30 to 60 days, just follow me + RT this post + reply with "SEO Stuff Cheat Codes" and I'll DM you. You must do all 3 for the DM. Alright, let's get into it. A recent Ahrefs study took a deep dive into data from 75,000 brands to see what actually determines visibility in AI Overviews. This data is the clearest evidence yet that SEO Stuff’s system (seo-stuff.com/gold-plan-pack…) is exactly what brands need to show up in AI search. Here are the key points: 1. AI visibility is built on content + mentions + authority. Ahrefs found the strongest correlations with brand visibility in Google’s AI Overviews were: Branded web mentions (0.664) Branded anchors (0.527) Branded search volume (0.392) Domain Rating (0.326) Number of referring domains (0.295) Backlinks (0.218) That’s your roadmap right there. AI search doesn’t just care about your site, but rather it also cares about how consistently your brand appears, how often others link to you, and how trustworthy your site structure looks when scanned by Google’s AI. That means content gets you mentioned, and backlinks get you trusted. You need both. 2. Mentions are multiplying visibility, but backlinks are what amplify it. Brands in the top 25 percent for web mentions earned 10x more visibility in Google’s AI Overviews than everyone else. But what do those top brands all have in common? They already had strong backlink profiles, which are still the foundation of Domain Rating and off-site authority. Ahrefs’ correlation data backs this up: DR (0.326) and referring domains (0.295) still matter. The more authoritative your domain, the more your mentions actually count. If content builds your reputation, backlinks validate it. AI systems now need both to decide who gets cited, linked, and surfaced in search results. 3. Paid ads and weak brands don’t make the cut. Branded ad traffic (0.216) and ad cost (0.215) barely correlated with AI mentions. You can’t buy your way into visibility. Google’s AI Overviews reference brands they trust, not brands that spend. And that trust is measured through your on-site clarity and off-site authority. That’s exactly what SEO Stuff (seo-stuff.com) builds. 4. Ahrefs’ study confirms the new order of operations: Structured content + consistent mentions + strong backlinks = AI search visibility, and SEO Stuff’s 3 core plans map perfectly to that formula: SEO Stuff Gold Plan: seo-stuff.com/gold-plan-pack… 10 long-form, AI-quotable articles + 3 DR50+ backlinks Structured with question-based headings, TLDRs, and schema Optimized for ChatGPT, Perplexity, Gemini, and AI Overviews Every article designed to get cited and indexed The Gold Plan is the foundation because it creates the content that gets surfaced and the backlinks that give it authority. SEO Stuff Premium Content Bundle: seo-stuff.com/premium-conten… 60 comparison-style articles targeting “best,” “top,” and “alternatives” keywords Each piece formatted for AI extraction (H2s, summaries, lists) Conversational phrasing and freshness signals built in Scales your brand visibility across Google and AI engines This is how you grow topical coverage and teach AI systems to associate your brand with your category. SEO Stuff Premium Backlink Bundle: seo-stuff.com/premium-backli… 3 DR50+ backlinks with analysis and reporting Strengthens the off-site authority AI models use to evaluate trust Reinforces your domain’s mentions and authority metrics Makes your brand “eligible” for inclusion in AI Overviews Content earns visibility and backlinks preserve it. Together they make your brand AI-recognized. 5. Ahrefs found that 26 percent of brands have zero presence in AI Overviews. Those that do are the ones consistently mentioned, linked, and cited, both on and off their own websites. That’s exactly what SEO Stuff (seo-stuff.com) is built to deliver. If you want some "unconfirmed" cheat codes for getting your brand cited in ChatGPT, Perplexity, and Google AI Overviews, just Follow + RT + reply “SEO Stuff Cheat Codes.” You must do all 3 for the DM.

@THArrowOfApollo RB2B

Adam Robinson fired us after we drove 42% of RB2B’s revenue. Six months later he rehired us. Now we’re using cold email to print him $500K/month in LTV. Here’s the full 2-year breakdown: PHASE 1: March 2024 - October 2024 RB2B went $0 → $4M ARR in 6 months. 42% of all revenue came from cold email. And our leads paid 2.7X more than their other channels. We weren’t just signing people up. We were signing up their best customers. Then Adam “paused” us while he reconfigured the product. I counted us as fired. PHASE 2: May 2025 - Now Six months later he rehired us. RB2B was at $5.3M ARR. Now they’re at $6.5M ARR. We’re currently generating between $130K to $500K in lifetime value every single month. (The video below explains why it’s a range) What most agencies would hide: We torched his infrastructure multiple times. Completely burned domains. I thought he was going to fire us again. But the conversion rate at $6.5M ARR stayed exactly the same as it was at $1M ARR. Still 1-in-250 emails signing people up. Adam being Adam, he asked me to document everything from the last 2 years. And I mean everything. The good, the bad and the ugly. So I built a 20-minute video + full strategy breakdown showing: - Exact attribution data from both phases - Why we got “fired” and what changed when we came back - How our leads convert 2.7X better than other channels - The email infrastructure disasters and how we rebuilt - Real client data most agencies could never share Comment RB2B below and I'll DM you the video link

@alexgroberman SEO Stuff Perplexity Cheat Codes"

Reddit just sued Perplexity and a bunch of scrapers (SerpApi, Oxylabs, AWMProxy). In the process they also exposed how SEO Stuff (seo-stuff.com) has been getting traffic + sales for customers from Perplexity (and ChatGPT) over the last 6 months. Now that it’s no longer a secret, let’s talk about it. But before we do... Want some cheat codes for landing in Perplexity + Google within 30 days? Just follow me + RT this + reply with "SEO Stuff Perplexity Cheat Codes" and I'll DM you. You must do all 3 for the DM. Alright, let's get into it. Reddit’s lawsuit claims Perplexity and its partners have been scraping Google’s search results to capture Reddit content indirectly, bypassing robots.txt restrictions. The most revealing bit from the complaint: Reddit apparently created a post that could only be crawled by Google, and within hours, Perplexity produced the content of that post. The only way Perplexity could have gotten that Reddit content, according to them, is if Perplexity and/or its co-defendants scraped Google SERPs and quickly incorporated that data into its answer engine. That tells us a lot. Perplexity (and the other AI search platforms) aren’t independently crawling the entire web in real time. They’re selectively pulling data from Google’s top results and pre-vetted authority domains, then embedding and scoring that content through their own ranking layers. Here’s what’s happening under the hood: Perplexity scrapes or queries Google SERPs to find fresh, high-ranking URLs. These URLs act as a feed for its own retrieval layer. That’s why visibility in Google directly translates to visibility in AI search. Retrieved content is chunked into embeddings. Perplexity (and ChatGPT) compute vector similarity between your content and a user’s query. Content with higher semantic alignment and clean structure (proper headings, concise answers, clear topic hierarchy) is prioritized. Domains already appearing in Google’s top results or linked from trusted sources receive a weighting boost. This explains why DR50+ backlinks still move the needle even in AI Search. Perplexity applies a time-decay variable similar to Google’s QDF scoring. The more recently a page was updated or re-indexed, the higher its visibility multiplier. That’s the exact system Reddit outlined in its lawsuit. And that’s exactly what SEO Stuff (seo-stuff.com) was built around. Gold Plan (seo-stuff.com/gold-plan-pack…) 10 AI Search-optimized articles + 3 DR50 backlinks per cycle. Articles are structured with high semantic density, clear markup, and snippet-first formatting, all factors that help AI crawlers parse and cite them quickly. PR backlinks from sites already appearing in AI search with legitimate, strong traffic. Premium Content Bundle (seo-stuff.com/premium-conten…) 60 topic-targeted, high-similarity articles updated on rolling cycles to stay inside AI visibility windows. This content is designed to appear in both Google’s index and Perplexity’s derivative index. Premium Backlink Bundle (seo-stuff.com/premium-backli…) 3 DR50+ contextual backlinks from domains already cited by AI assistants, reinforcing trust and entity alignment. Perplexity and ChatGPT are not indexing the entire internet. They’re indexing Google’s version of it, selectively filtered through authority, structure, and freshness signals. That’s why SEO Stuff (seo-stuff.com) customers who keep content fresh, semantically aligned, and consistently link-reinforced are the ones earning citations, traffic, and even direct conversions from AI search. If you want visibility in ChatGPT, Gemini, or Perplexity, you have to build for Google + AI ingestion, not Google alone. That’s been the SEO Stuff playbook since day one, and the Reddit lawsuit just happened to confirm it. Want some cheat codes for landing in Perplexity + Google within 30 days? Just follow me + RT this + reply with "SEO Stuff Perplexity Cheat Codes" and I'll DM you. You must do all 3 for the DM.

@scrapingdog very good scraping service

@samsonirl @AnnQuann Because they know Thailand won’t go further than the current long range exchange. None of side can afford attack with infantry

@AnnQuann Thailand had 10x the economy of Cambodia, mandatory military service for young men, and a monarchy that still holds all the power - why would Cambodia even try this?

2025 Cambodian–Thai Border Clash (Thread) Well, it's a stone's throw away from Vietnam so it naturally draws the attention of many Vietnamese including me. Let's get to it. (Will be updated as new stuff surface)

@GavinnyOriginal nie, na to treba pol naroda magorov.

Poslední dění v USA ukazuje, jak směšně jednoduché je během pár týdnů zničit něco, co systematicky budovaly desítky až stovky tisíc lidí celá desetiletí. Ať už to jsou mezinárodní vztahy, spojenecké vazby, USAID, věda, výzkum, školství... Stačí na to pár magorů bez morálky.

@Posledniskaut lol, Slovaci, vy ste tak tupy narod ?

Fico tvrdí, že v pozadí protestů na Slovensku jsou Gruzínské legie, které jsou součástí ukrajinské armády, konkrétně velitel Mamuka Mamulašvili. Propojení dokladuje třeba jeho fotografií s Lucií Štasselovou z iniciativy Mír Ukrajině. Dva roky starou fotografií z veřejné akce ke knize o něm od Tomáša Forróa. K fotografii Mamulašvili říká, že tu akci si sice pamatuje, ale tu paní vůbec nezná, netuší, kdo to je a dodává: "Dobře, pořád ještě nějak chápu, že se nás snaží obviňovat z rozpoutání protestů v Gruzii. Ale na Slovensku? To mi přijde opravdu zábavné a velmi zajímavé."

@henrythe9ths AI Guide

There's a shocking fact about AI that nobody tells you: You can catch up to the public AI research frontier in just 2 weeks. Yes, really. I've built a $150M annual revenue startup over the last 8 years and If I were to start a company today, I’d drop everything and go all-in on AI. But like many busy software builders, I felt lost—overwhelmed by the noisy, crowded and fast-moving modern AI landscape. And I wasn’t alone. So I spent my entire holiday diving deep into AI research—reading 30+ papers, watching hours of lectures, analyzing trends, and catching up to the research frontier. ✨ Here’s what I learned: - You don’t need months (or years) to catch up. - You don’t need a PhD or decades of ML experience. - You need fewer than 20 papers and 2 weeks to understand the major breakthroughs shaping AI today. It's because the technology is extremely nascent and most techniques that came before are no longer relevant: - ChatGPT is barely 2 years old and Transformers are only 7 years old. - Most game-changing discoveries happened within the last 4 years, driven by a few breakthrough ideas, scaling laws, and efficient matrix multiplication. The biggest secret? Many groundbreaking AI papers with thousands of citations are surprisingly simple and applied, like adding "let's think step by step" to the prompt, or simply asking the LLM over and over again to improve its answer (Self-Refine). I realized there are tons of founders and builders in the same boat—wanting to dive deeper into AI but unsure where to start. I've created an essential AI Guide that helped me catch up, in just 2 weeks, to the frontier of public AI research to figure out where the next opportunities and gaps were: - Curated list of only the most important papers - Simple explanations of key concepts - Clear pathway to understanding the frontier of modern AI It’s perfect for: - Founders expanding into AI - Builders wanting to innovate at the frontier of AI - Investors looking to separate the signal from the noise 👇 Want the full guide? - Like and Share this post - Comment "AI Guide" - I'll send you the complete guide (ps, I’m also teaming up with @VishalVasishth, co-founder of @obviousvc with @ev (focused on large-scale societal impact companies like Twitter, Medium, Beyond Meat), to host a small meetup to discuss what's working and needs to be solved in the AI stack in SF. Message me if you're interested)

@jakezward Can you share the link to verify what you are saying?

I've earned millions in SEO traffic using AI. This thread exposes my entire AI SEO workflow:

@MarekBulant @ostravanda_ 😀

@ostravanda_ Tohle vysvětluje pravdivost použití chemtrails k debilizaci nepřítele. Přirozené hlupství totiž nedosahuje takové úrovně. To musí být výrobek některé z podzemních laboratoří na Ukrajině placených Sorošem a řízených partou iluminátů.

Prosím, už dost! 😀 “V případě dárcovství krve je velká pravděpodobnost, že neočkovaný pacient může být příjemcem krve od očkovaného a tedy jsme ho očkovali proti jeho vůli.” MUDr. Peter Kotlár, vládní zmocněnec a poslanec NR Slovenska.

A client is looking for CKEditor replacement for their Rails apps, what are all the cool kids using these days?

@EffDeMulla @Kolowrat_ @front_ukrainian No. He is not against money

@Kolowrat_ @front_ukrainian Is PM Fico against this?

⚡️"🇺🇦Ukrainian armored vehicles" increases the supply of 122 mm high-explosive artillery shells - thousands of new shells were delivered today "Today, a batch of 122 mm high-explosive artillery shells, produced in partnership with one of the ammunition holdings of Eastern Europe, was delivered to Ukraine. Shots of this caliber are extremely necessary for the Ukrainian army - they are used daily for the D-30 and 2S1 guns," states General Director of Ukrainian Armored Vehicles LLC Vladyslav Belbas. According to him, the company managed to establish systematic serial production and significantly increase the production volumes of 122 mm artillery rounds.

My company, Calm Fund, has a big problem and I need your help. We are being sued by SureSwift Capital over the @FounderSummit event series we put on over the past few years. The main issue we have is the impending massive cost of doing discovery in a modern remote company that uses a lot of software, but first here is some context: SureSwift is a PE fund that acquires software companies. We signed a lightweight joint venture agreement to collaborate on the Founder Summit which was an IRL event in Mexico City, an online community and another event in North Carolina. After the final IRL event we terminated the joint venture. The events were by all accounts a huge success, and all in the joint venture posted a small net loss. Then we announced our intention to host our own slate of events in 2023, but ultimately had to cancel those due to budget constraints and reducing our team size. Then SureSwift sued us claiming that we had misappropriated the "intellectual property" and "trade secrets" like the "business model" and "virtual network connections" of the new-terminated Founder Summit JV for the now-cancelled Calm Summits. The suit is a matter of public record and I won't speak about it much further except to say that we deny the allegations and it is my strong belief we would win in a trial. The issue is that we are about to enter the phase pre-trial procedures called Discovery (or eDiscovery) where each side is obligated to share with the other side provide all sorts of their own documents and communications that could be relevant to the case. The big problem is that with modern remote companies using many collaboration tools to create emails, slack messages, docs comments, notion databases, etc the volume of data and the cost of discovery (collecting, reviewing, filtering and processing all these files) seems to have exploded. I've been hearing estimates of hundreds of thousands or more than $1m in costs just to comply with eDiscovery for a single case. That's an insane amount of money, especially relative to the stakes we're talking about in this matter. It puts a huge amount of pressure on folks like me to just settle since its not worth it to go to trial, spend even more money on pre-trial, and still have at least some small % chance of losing. As best as I have learned, in Delaware Chancery court it is also very rare to get your attorneys' fees reimbursed or be able to countersue for the costs when you win. So basically, the status quo situation in the American legal system is that if someone is willing to sue you and willing to spend a boatload of money themselves, they can force you to spend a boatload too (potentially all the way into bankruptcy) with no real defense or recourse unless you settle on whatever terms they dictate. When I asked my LPs for feedback or advice on this topic I was shocked by two things: 1. How many of them had dealt with a bogus lawsuit. 2. How essentially nobody had heard of any good ways to resolve it except paying out a settlement, even if you did nothing wrong. This is a totally insane state of affairs and has been eye-opening for me. Its also a huge risk for early stage founders who simply can't afford a big pay off settlement. I don't see any way I can bring myself to pay a settlement just because its cheaper than litigating the issues. Once you open that door, how do you avoid just becoming a piggy bank for anyone and everyone willing to bring a lawsuit against you? So I'd like to tackle this issue head on and try to find an affordable way to comply with eDiscovery so we can get to a trial and resolve the actual question. The reasons eDiscovery is so expensive don't seem intractable to me and there has got to be a way to do this that is at all proportional in costs to the actual matters of the case. But I need help. 1. Founders (or anybody really) if you have dealt with this issue before and found any kind of good solutions or can point to any good resources please reply below and amplify 🙌 2. I'm looking to collaborate with counsel or firms that have experience tackling this issue or want to. I know a big problem with this whole situation is that nobody on the legal side is particularly incentivized to help reduce costs on this matter, but believe me the entrepreneurial community will be forever grateful for help on this 🙏

Demystifying Decorators in TypeScript: Empowering Your Code with Elegance reliasoftware.com/blog/decorator…