Softaims

How Figma migrated from its own programming language, Skew to TypeScript They have been using Skew, a custom programming language, for their mobile rendering architecture. But recently, they switched to TypeScript, the standard language for web development. It wasn't an easy decision, though. Skew has its perks, but as Figma grew, it became more of a headache to maintain. Furthermore, it does not support other developer tools or communities outside of Figma. So, they had to migrate to TypeScript for new possibilities. The process was a bit tricky. They had to maintain performance, especially since they were moving from a language they had used for years. But they found good ways around it, like leveraging WebAssembly and swapping out key Skew components with stuff from their C++ engine. They initially followed JavaScript patterns for writing a transpiler. However, they faced some tricky challenges at the tail end, but somehow they managed to keep everything running smoothly. 🔹Avoiding JavaScript array destructuring increased performance by 25%. 🔹Ensured Skew's "devirtualization" optimization did not disrupt their code. 🔹Improved TypeScript symbol ordering in their transpiler for compatibility. The most impressive part They kept the whole team informed and ensured there were no major interruptions to work, even with all the technical stuff going on. They rolled out the migration in 3 phases, gradually shifting from Skew to TypeScript without causing any major problems. Phase 1: Develop the Skew transpiler and integrate it with the pipeline. Phase 2: Roll out TypeScript from Skew and fix type errors. Phase 3: Transition to TypeScript, remove Skew, and notify developers. This method ensured control over Skew, a smooth transition, and rapid issue resolution during the rollout. Figma reaped several benefits from migrating all Skew code to TypeScript: ✅ Modernized a key codebase. ✅ Improved integration with internal and external code. ✅ Boosted developer efficiency. ✅ Exploring additional benefits, like easier package management & more. Figma's move to TypeScript helped it upgrade its codebase and position itself for future growth and innovation in an ever-changing tech landscape. What's your take on it? Share your thoughts below 👇 ------ Enjoy reading our content? Follow Softaims for more! Have an idea? Let's connect and build your dream startup 📈 📍Visit us: thesoftaims.com #softaims #theSoftaims #figma #softwaredevelopment #coding #programming #developers #build #typescript #javascript #engineering #tech #technology #hiredevelopers #skew #startup #website #ecommerce #innovation #c++ #backend #WebAssembly

Slack is primarily based on a monolithic service - The Webapp They use ReleaseBot to automate the deployment of updates, ensuring a smoother, faster, and more reliable process. With great size comes great responsibility, particularly when it comes to implementing changes. You see, in software development, deploying updates can be challenging. While many people talk about the benefits of constant deployment, dealing with 150 changes on an average day is a completely different story. Slack deploys the Webapp repository 30-40 times daily, with an average deployment size of 3 PRs, all managed by a ReleaseBot that automates deployment tasks. ReleaseBot effectively tracks deployments by detecting anomalies like z scores and dynamic thresholds. But it wasn't always this seamless. Back in the day, before ReleaseBot, there were Deploy Commanders (DCs). These guys would guide the Webapp through its deployment steps, keeping a full eye on dashboards and executing manual tests along the way. Yet there was a problem. Despite their best efforts, the DCs often found themselves second-guessing their decisions. It's hard to feel confident when you're navigating a system as complex as Webapp, especially when you're on rotation with hundreds of other developers. So, engineers at Slack's Release Engineering team came up with a solution - ReleaseBot. They designed this bot to automate deployments, using advanced anomaly detection techniques like z scores and dynamic thresholds. 📝 z score for a particular data point is ((data point – mean) / standard deviation) They use z scores as "high confidence" signals to spot important changes and urgency levels with emojis. They also monitor static thresholds, considering them "low confidence" to keep operations running smoothly, ensuring normal fluctuations do not interfere with deployments. The benefits Slack gains from using ReleaseBot are manifold. - Ensures smoother, faster, and more reliable deployment processes - Operates 24/7 to continually deploy new builds - Provides higher-level, clearer "go/no-go" signals to deployment commanders. - Utilizes anomaly detection and monitoring to catch issues faster and more consistently than human counterparts. - Enhances efficiency and reliability. It is scary to automate deployments. But, with a careful, iterative approach, you can alleviate your fears and achieve rapid deployment while reducing the risk of errors. Don't let fear hold you back. By understanding the nuances of deployment monitoring, you can also have a smoother, faster, and more reliable deployment process. -------- Enjoy reading our content? Follow Softaims for more! Have an idea? Let's connect and build your dream startup 📈 📍Visit us: thesoftaims.com #softaims #theSoftaims #slack #softwaredevelopment #coding #programming #devlife #developers #build #ReleaseBot #Webapp #monolithic #engineering #tech #technology #hire #devteams #startup #startups #business #ideation #innovation

Thanks for hosting such a wonderful event, Stack360. We had a great time! You guys rocked Tekken 7. Proud of our team for bringing their best in table tennis. Can't wait for the next showdown! #indoorgames #softaims #Stack360 #videogamelife #goodtimes #networkingevent #techevent #communitybuilding #itcommunity #innovation #thesoftaimslife #softaimsculture #competition

Slack is primarily based on a monolithic service - The Webapp They use ReleaseBot to automate the deployment of updates, ensuring a smoother, faster, and more reliable process. With great size comes great responsibility, particularly when it comes to implementing changes. You see, in software development, deploying updates can be challenging. While many people talk about the benefits of constant deployment, dealing with 150 changes on an average day is a completely different story. Slack deploys the Webapp repository 30-40 times daily, with an average deployment size of 3 PRs, all managed by a ReleaseBot that automates deployment tasks. ReleaseBot effectively tracks deployments by detecting anomalies like z scores and dynamic thresholds. But it wasn't always this seamless. Back in the day, before ReleaseBot, there were Deploy Commanders (DCs). These guys would guide the Webapp through its deployment steps, keeping a full eye on dashboards and executing manual tests along the way. Yet there was a problem. Despite their best efforts, the DCs often found themselves second-guessing their decisions. It's hard to feel confident when you're navigating a system as complex as Webapp, especially when you're on rotation with hundreds of other developers. So, engineers at Slack's Release Engineering team came up with a solution - ReleaseBot. They designed this bot to automate deployments, using advanced anomaly detection techniques like z scores and dynamic thresholds. 📝 z score for a particular data point is ((data point – mean) / standard deviation) They use z scores as "high confidence" signals to spot important changes and urgency levels with emojis. They also monitor static thresholds, considering them "low confidence" to keep operations running smoothly, ensuring normal fluctuations do not interfere with deployments. The benefits Slack gains from using ReleaseBot are manifold. - Ensures smoother, faster, and more reliable deployment processes - Operates 24/7 to continually deploy new builds - Provides higher-level, clearer "go/no-go" signals to deployment commanders. - Utilizes anomaly detection and monitoring to catch issues faster and more consistently than human counterparts. - Enhances efficiency and reliability. It is scary to automate deployments. But, with a careful, iterative approach, you can alleviate your fears and achieve rapid deployment while reducing the risk of errors. Don't let fear hold you back. By understanding the nuances of deployment monitoring, you can also have a smoother, faster, and more reliable deployment process. -------- Enjoy reading our content? Follow Softaims for more! Have an idea? Let's connect and build your dream startup 📈 📍Visit us: thesoftaims.com #softaims #theSoftaims #slack #softwaredevelopment #coding #programming #devlife #developers #build #ReleaseBot #Webapp #monolithic #engineering #tech #technology #hire #devteams #startup #startups #business #ideation #innovation

Best practices for those planning to implement Kafka: 1/ You need real data for accurate performance testing 🔹Performance tests cannot be done without real data from producers and consumers. 🔹High and low Watermarks should be set based on actual testing rather than theoretical values. 2/ Match partitions per consumer for parallelization 🔹Each consumer should have the same amount of partitions. 🔹The number of consumers should match the number of partitions. 3/ Batching correctly is crucial for Kafka scaling 🔹Consider the type and amount of batching for optimal performance. 🔹Spool up producers to put on large batches for efficient processing. 4/ Testing message throughput and average message size is crucial to optimizing Kafka performance. ------------ Enjoy reading our content? Follow Softaims for more! Have an idea? Let's connect and build your dream startup 📈 📍Visit us: thesoftaims.com #apachekafka #streamingdata #realtimedata #datamanagement #datastreaming #softaims #developers #webdev #hire #coding #concurrency #performance #efficiency #tech #technology #programming #techsolutions #hiredevelopers #digitaltransformation #webservices #techworld #innovation

Identify the patterns in your thought process, especially about money. It's NOT about chasing the dollar; it's about understanding. Here are 6️⃣ steps that can help you change your mindset: 1/ Rethink money Money should always be a byproduct, never the end product. Our family backgrounds significantly shape our perceptions of money, especially if we grew up in a low-resource family, leading to the belief that earning money is extremely difficult. Is money what you work for 24/7? Can you earn it through hard work, or is it what one inherits from ancestors? But Forbes reports indicate that most billionaires are self-made, and their numbers are on the rise. 2/ Create value Producing value on a large scale can lead to earning money on a large scale. In today's capitalist world, you earn the money you create value for. Create value by utilizing skills and interests to build something valuable. The value of a product or service is determined by customer's willingness to pay. For example, a bottle of water in a desert is worth much more than one in a grocery store. 3/ Learn on someone else's dime The best way to proceed is to learn from experienced people and then start your own thing. You don't have to learn everything on your own; you may work with others and learn from their experience. "History does not repeat itself, it does rhyme." 4/ Solve problems Big companies like Amazon and Tesla are making billions by solving real problems. If you want that much money, stop chasing it and solve problems. The money will follow. They succeeded financially by solving critical issues, proving that success often comes from solving legitimate problems. 5/ Learn from failure Failure is a part of the journey. Everyone fails at some point in their journey. What matters most is that you learn from your mistakes and move on. Your first ideas might not work out. It's okay. Sometimes, your later ideas do better. For example, Bill Gates' first business, Traf-O-Data, failed. But he didn’t stop there; he learned from his mistakes and later found one of the world's most successful companies, Microsoft. 6/ Aim for freedom The true value of money is the freedom it provides. Hard work and dedication are required to make it to the end, but those who reach the end of the road will experience wealth beyond their imagination. Money provides freedom and options, but it alone doesn't bring happiness. It's about mindset and choices. Money can limit you, but not needing it empowers you to take control. tl;dr: Seeking financial freedom means focusing on finding solutions to problems rather than chasing money. Value your time and not waste it on things that won't help you live the life you want. ------------ Enjoy reading our content? Follow Softaims for more! Have an idea? Let's connect and build your dream startup 📈 📍Visit us: thesoftaims.com #money #mindset #softaims #hiredevelopers #tech #software #development #startups #digitalsolutions

Lack of discipline in software documentation has several negative effects on the overall development process & project success. Think of documentation as a user manual for your code. It saves you and your team a ton of time and resources. Here are just 5 benefits you can't ignore: 1/ Save time & reduce errors 2/ Lower maintenance costs 3/ Enhanced collaboration 4/ Accelerate developer onboarding 5/ Better/more reliable code Keeping up with code documentation❓ ✅Follow Softaims for more content like this! 📍Visit us: thesoftaims.com #codedocumentation #documentation #savetime #transparency #quality #trust #confidence #softwaredevelopment #codingtips #bestpractices #softaims #efficiency #teamwork #tech #success #documentationmatters #techskills #codingcommunity

Craig Clemens sold over $1 billion of products online with his marketing strategies. Guess what? We've cracked his 7 Human Hijacking Strategies—the magic behind the scenes. These aren't just tips, they're billion-dollar secrets. Don't miss out! 1️⃣ Make it about customers & the benefits they going to get Nike's "Just Do It" campaign is a prime example of a customer-focused advertising campaign that focuses on the benefits of its users. 2️⃣ Make it an event When you turn your marketing into an event, you boost brand visibility, grab attention, & create stronger relationships with your audience. A stellar example of marketing as an event is Apple's product launches. Apple hosts well-planned keynote presentations rather than quietly releasing new products. You may have watched the recent Vision Pro Headset release live stream. 3️⃣ Powerful demonstration When you demonstrate your product or idea effectively, you don't just tell people what it can do—you show them. Think about how the product will benefit the user and solve a problem—& what makes it better than other products. A powerful demonstration speaks louder than words. Elisha Otis proved this by showcasing his elevator's safety features at the Crystal Palace in NYC. 4️⃣ Show unquestionable proofs The best ally you have in sales is proof. When you show proof that can not be questioned, people will trust and believe in your product/idea. Look at Tesla's Autopilot safety features: They prove beyond a doubt that their self-driving technology is safe and reliable by using rigorous testing, real-life examples, & statistical data. It also shows that they care about the safety of people, which makes people more likely to buy from them & builds trust. 5️⃣ Change a daily behavior Make your product or service part of your audience's daily routine to give them a sense of satisfaction and improvement they will want to talk about. If you change your customers' daily behavior, it has a viral effect. Find ways to make your product part of your target audience's daily lives, forming positive habits that lead to satisfaction and organic word-of-mouth marketing. 6️⃣ Sell the dream Many in the sales industry often put it this way: “Nobody who bought a drill actually wanted a drill. They wanted a hole.” People have dreams, big or small. When your product aligns with those dreams, it becomes irresistible. 7️⃣ Help clients feel superior People often want to stand out or feel special. It's a basic human desire, & you can leverage it to make your product or idea more appealing. Think about luxury fashion brands like Chanel or Gucci. These brands don't just sell accessories, they sell a lifestyle and a sense of exclusivity. Marketing plans can either make or break a company. Even if you only use one of these strategies, it has the potential to completely transform a company & increase its success if done correctly. #marketing #softaims #marketinggoals #brandbuilding #marketingtips

How Airbnb uses ElastiCache Redis to store +237 terabytes of data Airbnb does not just use Redis as a persistent data store. They use it as a really fast in-memory data store. Before migrating to ElastiCache, Airbnb's Redis setup was self-managed and consisted of two main architectures: ⒈ SmartStack Redis ⒉ High Availability (HA) Redis SmartStack Redis architecture featured primary and replica instances but faced scalability issues and lacked support for sharding and encryption. On the other hand, HA Redis architecture offered auto-failover and multi-AZ support but required significant operational overhead to manage Redis Sentinel clusters. There was neither support for sharding nor encryption support, and with this architecture, they started using 2 Ton Epoxy, which posed many issues, such as restricted commands. Why did they choose to use ElastiCache? Because → The current approach didn't work → Difficult to manage → Poor user experience → Doesn't scale → The EC2 instances or instance types they had to use in many clusters were expensive. They found several features in ElastiCache Redis to be especially useful: 🔹Key-value store: Swift in-memory data storage within the cloud, serving as a versatile database, cache, message broker, or queue. 🔹Fully managed & secured: AWS handles hardware, software, setup, configuration, monitoring, recovery, and backups. 🔹Secure & compliant: VPC ensures cluster isolation alongside encryption for data at rest and in transit. 🔹Highly accessible & dependable: Leveraging read replicas, multiple primaries, and multi-AZ setups with automated failover. 🔹Easily scalable: Supporting clusters with up to 6.1 TiB of in-memory data and offering read scaling with replicas, as well as write and memory scaling through sharding, scale-out, or scale-in. Amazon ElastiCache is a serverless, Redis, and Memcached-compatible caching service that delivers real-time, cost-effective performance to modern applications. ElastiCache can scale to hundreds of millions of operations per second with microsecond response times while maintaining enterprise-grade security and reliability. 👍Making a serverless cache that is always available and can grow instantly to meet demand takes less than a minute. ✅Implement real-time session stores to quickly manage ephemeral session data, facilitating quick personalization for gaming, e-commerce, social media, and online applications. Follow Softaims for more content like this! #amazon #softaims #ecommerce #elasticache #Airbnb #Redis #scalability #sharding #tech #softwaredevelopment #scale #programming #cloud #software

How Airbnb uses ElastiCache Redis to store +237 terabytes of data Airbnb does not just use Redis as a persistent data store. They use it as a really fast in-memory data store. Before migrating to ElastiCache, Airbnb's Redis setup was self-managed and consisted of two main architectures: ⒈ SmartStack Redis ⒉ High Availability (HA) Redis SmartStack Redis architecture featured primary and replica instances but faced scalability issues and lacked support for sharding and encryption. On the other hand, HA Redis architecture offered auto-failover and multi-AZ support but required significant operational overhead to manage Redis Sentinel clusters. There was neither support for sharding nor encryption support, and with this architecture, they started using 2 Ton Epoxy, which posed many issues, such as restricted commands. Why did they choose to use ElastiCache? Because → The current approach didn't work → Difficult to manage → Poor user experience → Doesn't scale → The EC2 instances or instance types they had to use in many clusters were expensive. They found several features in ElastiCache Redis to be especially useful: 🔹Key-value store: Swift in-memory data storage within the cloud, serving as a versatile database, cache, message broker, or queue. 🔹Fully managed & secured: AWS handles hardware, software, setup, configuration, monitoring, recovery, and backups. 🔹Secure & compliant: VPC ensures cluster isolation alongside encryption for data at rest and in transit. 🔹Highly accessible & dependable: Leveraging read replicas, multiple primaries, and multi-AZ setups with automated failover. 🔹Easily scalable: Supporting clusters with up to 6.1 TiB of in-memory data and offering read scaling with replicas, as well as write and memory scaling through sharding, scale-out, or scale-in. Amazon ElastiCache is a serverless, Redis, and Memcached-compatible caching service that delivers real-time, cost-effective performance to modern applications. ElastiCache can scale to hundreds of millions of operations per second with microsecond response times while maintaining enterprise-grade security and reliability. 👍Making a serverless cache that is always available and can grow instantly to meet demand takes less than a minute. ✅Implement real-time session stores to quickly manage ephemeral session data, facilitating quick personalization for gaming, e-commerce, social media, and online applications. Follow Softaims for more content like this! #amazon #softaims #ecommerce #elasticache #Airbnb #Redis #scalability #sharding #tech #softwaredevelopment #scale #programming #cloud #software