The Box Model in CSS 🎨 ✧ In HTML, all elements are considered as a box. A box comprises padding, border, and margin in addition to its actual content. A thread 👇🏻

I built this thing called Clicky. It's an AI teacher that lives as a buddy next to your cursor. It can see your screen, talk to you, and even point at stuff, kinda like having a real teacher next to you. I've been using it the past few days to learn Davinci Resolve, 10/10.

Many are asking if they can move their Next.js projects from Vercel to Replit? Yes! - Go to Import then GitHub - enter repo URL - Agent will take care of the rest It will set up the dev and deployment environments! For companies moving lots of work, happy to help discount.

Every billion-dollar startup you admire started the EXACT same way: localhost:3000 Not AWS. Not Kubernetes. Not microservices. Not Docker containers. Just a developer hitting "npm start" on their laptop at 2 AM.

Here’s how you can use famous AI tools to deploy your products faster: 1) Use Perplexity AI for research. 2) Use ChatGPT to turn that research into a clear plan. 3) Use v0 .dev to design your product 4) Use Cursor to write the code. 5) Use Claude to fix bugs. All of this? Costs about $100/month.

Claude 3.7 Sonnet Replit Agent v2 It's been a HUGE week for AI-powered development. Two major releases just dropped that are changing how indie hackers & solo founders build products. Here's my experience with both: 1/ Anthropic's Claude 3.7 Sonnet Just tested this and I'm genuinely blown away. This isn't just another incremental AI update - it's redefining what's possible for developers. What's new: - The new "extended thinking mode" solves complex problems that stumped earlier models - Generating code that works the first time - Command-line interface that feels like having a senior dev on your team The most impressive part? It doesn't just write code - it writes *good* code with proper architecture and clean patterns. You can create entire Next.js applications in one shot. 2/ Replit Agent v2 Replit also just launched their second-gen AI coding assistant, and it complements Claude perfectly. The standout features: • Build full applications from plain English commands • Real-time design preview as you describe what you want • An intuitive workflow for non-tech founders At $15/month (Core plan), it's saving weeks of development time and thousands in contractor costs. What this means for indie hackers/solo founders The barrier to entry for building software just dropped dramatically. Ideas that would have taken months can now be prototyped in days. MVPs that cost $20K can now be built for a fraction of that. Features that required deep technical expertise are suddenly accessible to non-coders. We are seeing the advancement of real-time software development. 3/ My take These tools aren't replacing developers - they're supercharging them. The real value is in the problems you solve and the products you design. The winners will be the founders who learn to leverage these tools effectively while focusing on what truly matters: solving real customer problems. People are building crazy stuff. Need proof? Check out what @SullyOmarr built with Claude 3.7 Sonnet using this simple prompt: "please create me a saas landing page template designed extremely well. create all components required"

If you’re into health and have NoCode skills, this is a great opportunity

Your Conversion Rate is the outcome of 2 crucial steps: 1. A strategic approach to turning visitors into customers 2. Consistent optimization to maximize conversions at each stage For the first step, you need to: - Decide WHAT actions you want users to take - Ensure your product truly solves their problems - Make your value proposition crystal clear Track these 3 key points in your SaaS pipeline. First step is easier because YOU control the user experience. But it gets trickier with the second step: You need to turn potential into paying customers. You need to: - Go from "I want X% to convert" - To "X% of my users are converting" You don't decide anymore, YOUR USERS decide. That's why the second step is harder and ongoing: - Optimize your website and onboarding - Use content to showcase value - Stay consistent with your messaging - Keep testing and improving (even when it's frustrating) Keep refining your funnel to achieve the conversion rates you desire. Remember: - Average B2B SaaS trial-to-paid: 25-30% - Rates vary widely (17.4%-51%) based on factors like: • Industry and product complexity • B2B vs B2C • Marketing channels • Subscription model • Product maturity Higher conversion rates = more customers = lower CAC

Just read about @retool's fascinating journey on Reddit: • Founded in 2017 by David Hsu, 6 months after graduating from Oxford • Solved the problem of slow internal tool development • Started with YC, focused on staying "default alive" • Find market fit fast by changing messaging, not product • Grew to $500K revenue in first 9 months through cold outreach • Public launch in 2018 with $2M ARR and 40 customers By 2023: $93.5M revenue, 500,000 customers including DoorDash, Mercedes-Benz Last raised $45 million at a valuation of $3.2 billion in 2022. Key takeaways from Retool's journey: → Product-led growth freemium model → Heavy focus on customer success and rapid problem-solving → Sophisticated outbound sales inbound lead optimization → Built strong community, fueling organic growth It's impressive how they turned a "boring" problem into a $3.2B company. Shows the power of finding the right product-market fit, perfect timing, and relentless execution.

Most of us struggle to build a culture of excellence. Here's how Netflix builds a winning team: Netflix, with over 25 years in the industry has built a strong, unique culture. Netflix's CTO, Elizabeth Stone, leads the tech team. She is the first economist to hold this position at a Fortune 500 company. Her team has built a streaming platform serving 230M subscribers globally. Here's how Netflix recruits, manages, and empowers its teams to sustain its culture: Hiring Netflix offers salaries 10–20% higher than the average in the industry to attract top talent. They customize compensation based on individual market values. Ensuring employees feel valued without the constraints of long-term compensation packages. This approach keeps the talent bar high and helps Netflix attract the top 1% of candidates. Management Netflix uses the "keeper test" where managers ask if they’d fight to keep an employee. Their 8% turnover rate is well below the industry average and helps them maintain a high talent density. Netflix gives continuous, direct feedback to employees instead of traditional performance reviews. This makes employees 15% happier. Feedback at Netflix is direct and candid, yet supportive. Managers give feedback one-on-one, which has helped them keep 12% of their employees. Culture They focus on what matters most. And produce over 1,500 hours of original content each year. They provide significant autonomy to their 11,300 employees. This freedom, along with a lot of responsibility, has made the team 30% more productive as a whole. They also prioritize smart work over long hours, with an average workweek of 50 hours. Netflix revises its culture deck once a year. They pay above-market salaries to attract and retain talent. When hiring, they look for people who will improve the entire team rather than just fill a role. The culture has a lot of expectations, which can be stressful. Open communication is meant to help with this. What's your take? --- That’s it for now! If you found value in this post, like and share it with your audience. Follow me @umar482 for more!

@gregisenberg built a 7-figure business using Reddit and AI tools. The way he does it is truly mind-blowing. Here’s his process 🧵

Why Notion horizontally re-shard its Postgres database? In 2023, Notion's team switched the live database PostgreSQL cluster from 32 to 96 machines with no downtime. Notion's database architecture has evolved several times to keep up with the rapid growth of its user base. Initially, the company relied on a single, large Postgres database on Amazon RDS. As usage increased, the database was horizontally sharded to handle the increasing load. The 32-machine space shard served well for a while. However, as Notion continued to grow, several issues emerged: Problems they were facing • 90% CPU utilization at peak traffic • Near-full disk bandwidth (IOPS) utilization • Connection limits in PgBouncer With 5.8 million potential user firms and an anticipated new-year traffic spike, the company needed a swift, careful solution to maintain performance and availability. After evaluating multiple solutions, they decided to add More Databases! They chose to scale horizontally by tripling the number of instances from 32 to 96, allowing for load distribution and improved performance. 🔹Implementation steps 1/ Setting up the New Shards They created a new database cluster using smaller instance types and disks. ⇢ Created 96 new database instances ⇢ Reduced from 15 to 5 logical schemas per database ⇢ Used Terraform for automated provisioning 2/ Synchronizing Data They used Postgres logical replication to copy historical data and continuously apply new changes. ⇢ Created 3 Postgres publications per existing database ⇢ Set up subscriptions on new databases to consume publications ⇢ Skipped index creation during initial copy to speed up the process 3/ Verification They run tests to ensure the integrity and performance of the data. ⇢ Split PgBouncer into 4 groups to manage connection limits ⇢ Gradually migrate traffic to avoid overwhelming databases 4/ Testing Implemented "dark reads" to compare results between old and new databases ⇢ Limited comparisons to queries returning up to 5 rows ⇢ Implemented a 1-second delay in NodeJS web server using `setTimeout` to enable replication catch-up 5/ Failover They executed a careful failover process to switch traffic to the new shards without downtime. ⇢ Paused traffic to existing database ⇢ Updated PgBouncer shard mapping ⇢ Resumed traffic to new shards ⇢ Implemented reverse replication for rollback capability ✅ By extending horizontal sharding, Notion's system scaled to handle increased load, providing a better user experience. ✅ The new CPU and IOPS utilization stays around 20% during peak traffic, so there is much room for growth. Source: notion.so --- That's all for now! If you're finding value in this post, like and share it with your audience. Don't miss out on more insightful content — follow me @umar482

How engineering leaders build High-Performance teams? Building a high-performance team demands sharp focus & top Engineering Leaders know this. Here’s their blueprint 🧵↓

Netflix has around 2800 Java applications & around 1500 internal libraries. Netflix is primarily a Java-based company. Uses Java for its backend across its various applications. Including - internal tools - film production software - streaming. Over the years, they have continuously evolved their Java usage to meet the demands of their large-scale video streaming platform. They previously used custom frameworks and legacy Java EE libraries. How Java has evolved at Netflix: Netflix's journey to adopt different technologies shows their quest for efficient, scalable solutions. It all started nearly seven years ago when Netflix started using Groovy in a microservices setup, where each service handled specific tasks and data. They used Groovy scripts and a backend-for-frontend system for better data retrieval and integrated RxJava and Hystrix for fault tolerance and concurrent operations. However, managing scripts and reactive programming was difficult, leading to data fetching problems with their REST APIs. So, Netflix switched to GraphQL Federation, combining microservices into Domain Graph Services (DGSs) with a unified GraphQL schema. This change streamlined data fetching, removed the need for separate backends, and reduced reliance on RxJava and Hystrix. GraphQL’s schema-based querying provided accurate data tailored to UI needs, making development more efficient. 🔹Java at Netflix Netflix relies heavily on Java to support diverse functions, with 2800 microservices and 1500 internal libraries. Their Java setup includes the Azul Zulu JVM, an OpenJDK build. They use Gradle with Nebula plugins for version control and dependency management, and IntelliJ with custom plugins for development workflows. The transition from Java 8 to Java 17 marked a significant milestone for the company. Before, they were stuck on Java 8, even though Java 11 and 17 were available. The delay was due to older microservices running on a custom stack with outdated libraries. When they finally upgraded to Java 17, they saw a 20% improvement in CPU usage with the G1 garbage collector on Java 17. For their proxy, Zuul, they use the Shenandoah garbage collector because of its low pause times. They have begun testing Java 21 as well, as it is already out. Moving to Java 21 from Java 17 is easy and brings exciting features. Virtual threads offer better scalability, and the enhanced ZGC garbage collector works for more use cases. Java 21 also introduces data-oriented programming with new language features, making Java even more powerful. 🔹Spring Boot 3 Netflix has successfully migrated from its legacy Java stack to Spring Boot. It was challenging, but it unified their development process. All new applications are developed using Spring Boot, and Netflix has developed Spring Boot Netflix, which incorporates gRPC, Spring Security, and observability features. The switch also enabled the transition from Java 8 to Java 17, with plans for Java 21 . From in-house frameworks to modern microservices using RxJava and GraphQL Federation, Netflix's Java stack has improved scalability and efficiency with each adaptation. --- That's all for now! If you're finding value in this post, like and share it with your audience. Don't miss out on more insightful content — follow me @umar482

I've written a lot about founders, but here are the top 10 stories you must read. These founder stories will blow your mind & inspire you to work harder. From garages to billion-dollar companies, these are pure gold. Don't miss out — bookmark for later 🧵

I recently read an article on how @37signals handles internal communication & it’s quite enlightening. Here's how they ensure efficiency in their communications: 🧵