Sam Williams’ Post

Wanna get ready for next startup pitch or demo, but needs to deploy your product comprise of flutter + backend apis, then this architecture can help you go live cost-effectively. For Production environment, we should use application load balancers with Cloud CDN & SSL support. #startups #flutter #googlecloud #devops #flutterdevelopers #developers

A couple of months ago, #CrowdSnap joined the Microsoft for Startups program, which helped resolve many of our infrastructure challenges. I want to share some easy-to-use services that have been immensely beneficial in our early stages and have contributed significantly to our growth. 1) #Heroku - PAAS platform that never let me down. Incredibly fast deployments and uncomplicated CI/CD. Not free anymore but very reasonable pricing. Value for money! 2) #NodeJS / AngularJS - Extremely handy in creating serverless architecture and Single page apps. Highly recommend this combination for any scalable app 3) #TiDB - There are many things I love about them. 5GB free to start with, multiple regional clusters, super easy to setup, 100% MySQL compatible and scalable yet feels very light weight. Above all their support team cares for your app more than you do, trust me! 4) #Cloudinary - 10 GB storage to start with. You can use them for trusted CDN and very supportive support team Building a product is not easy. You need many combinations to play together to till you find a correct formula. I highly recommend above technology service providers for early stage tech startups.

❗Notify Me Weekly Update ❗ Hi all 🌻, here is a Notify Me update for this week: 📣 Updates: 🔹We have 1973 users (+31 since the last update) 🔹Number of paying users: 4 (+0 since the last update) 🔹MRR: $24.96 🔹Monthly costs on GCP: ~$67.11 (it still needs time to adjust the projection; our cost should be about 50% down by the end of the month); Hetzner: $13.18 (this includes everything we host there, some things unrelated to Notify Me) 🔹We finally migrated prod scrapers to Hetzner; this should reduce our costs by about 50%. We are currently running 8 scrapers non-stop for $5.99 A MONTH. Google Cloud setup was capped at 7 scrapers max, and all of them were serverless - which meant slower startup time and slower responses in general. 🔹False positive detection is in development, and we should hopefully be able to reduce false alerts which happen when content is changed (e.g., “4 days ago” → “3 days ago” type of changes) 🔹Traffic from Reddit is still going strong, and that’s reflected in 31 new users this week. Our SEO efforts are also paying off with the increased impression count. 🔹Competition Tracker will be ready for testing by the end of this week. ✅ Do’s: When building new products, speed is of the essence, of course, but at least try to set up a good foundation - especially with deployment automation and somewhat good programming practices. It will take you 3-4 days instead of a couple of hours, but it will pay off in the long run. ❌ Don’ts: Don’t underestimate the importance of formal education. A bunch of this might seem irrelevant or straight up useless, but a few years in the future you’ll be in a situation where some of that knowledge really comes in handy. Worst case - it will show you where to look for the answer. ❓ What is our goal for next week: 🔹One more paying customer 🔹Improving scrapers 🔹Having Competition Tracker completely operational I hope this was helpful, and I wish you an awesome week 🙌 #buildinginpublic #indiehacking #startups

Is This the Most Overhyped Tech Stack of 2024 for Solo SaaS Founders?😮 As a solo SaaS founder, I've been keeping an eye on the latest trends, and I have some thoughts on what's being touted as the "must-have" tech stack for 2024: 👉 Next.js 14: I love this framework, actually. It's powerful and versatile, but it's not the only option out there. 👉 Clerk or Kinde for Authentication: Why trust a third party with your users' login credentials? There are more secure and reliable ways to handle authentication in-house. 👉 Stripe as a Payment Processor: Yes, it's ubiquitous, but it's not the only option. There are other payment processors that might better suit your specific needs. 👉 ShadCN for UI: Overrated and overused. It even looks ugly if not customized properly. There are other UI libraries that can offer more flexibility and aesthetic appeal. Chakra UI, and DaisyUI look way better. 👉 Vercel for Deployment: This could be a risky choice. If you get hit by a DDoS attack, server costs can skyrocket into the thousands. Consider alternative deployment options that offer better cost control. Can someone explain this weird trend? Why are we pushing solo founders towards these particular tools? Here's things how I'd do differently when building a SaaS: - Use Auth0 or custom authentication: 🔐 For secure and customizable authentication solutions. - Mollie for Payments: Way easier to setup. Like why is everyone using stripe like it's the only option? - AWS or DigitalOcean: 🌐 For deployment, offering scalable and reliable infrastructure with a wide range of services. Let me know what you think! Also don't forgot to follow for moor SaaS startup tips. #nextjs #solofounder #progamming #react #developer #github #saas

AWS Amplify automatically sets up a GraphQL API (both server and client) using a Typescript file that keeps things type safe. That is hugely helpful, and saves a team a lot of time trying to handle infrastructure. Having your infra code integrated into your app repository, in the same language as the rest of the app, makes it less of a black box separating app and infra teams. GraphQL also means access to subscriptions, unlike REST, if those are needed. #aws #startup #tech #backend #frontend #api

A $96,280 bill. Do you see yourself, your project, or your team in the picture below? (image clipped from Jeff's video) I just watched a great, insightful video by YouTuber Jeff Delaney about the recent costs and consequences for real developers (cara.app) using tools like Vercel, Netlify, Heroku, etc. Check it out here: https://lnkd.in/ehNvhr5x Here's my advice and words of warning. While I appreciate what these tools offer—quick setup and deployment capabilities—they can become a cost trap as your product scales, especially for new developers and startups. Remember: These tools are excellent for getting started, but to scale properly and avoid hefty bills, you'll eventually need to pivot. If you're struggling with the cost of running your applications in production, let's chat. We have a track record of reducing infrastructure costs on AWS by over half. Have you used these tools before? What was your experience? Should these tools be held more responsible for the infinite scaling they provide to developers? Let me know your thoughts below!

99% of startups would be fine starting with Rails and two EC2 instances. Look I love designing robust scalable things as much as the next techie, but you shouldn’t start there. I’m always amazed when I talk to people in very early stage ventures and hear about their web scale setups vs single digit user counts. I’ve also seen too many teams struggling to roll their own solutions to things that full stack web frameworks do out of the box (auth, orm, api, etc). Experiments that could be done in hours drag on for weeks. I should start a consulting business that’s just replacing whatever’s there with a symfony app.

It’s not hard for experienced engineers to start something scalable. Many people do it all the time, without wasting time. Don’t mistake scalability for cool-but-unnecessary, time-consuming tasks. Complexity and over-engineering are not the same as scalability. More than 99% of startups don’t need the massive scalability like big tech companies, but they should be able to handle at least double or triple of thier own peak load.

Building and working with startups has been a crazy journey, and I have learned a lot about building a product from 0 to 1. Here are a few tips on How to build a Product from Scratch: - Write your ideas down but don't over ideate. - Don't spend huge capital on UI at the start. Head to V0.dev and get some free UI designs. - Setup a project in next.js and use app router iykyk. - Create a git repo (makes deployment insanely easy with AWS). - Use AWS Amplify to deploy UI. - Buy a domain and email service from none other than Hostinger. - Cognito for User management. - Backend hosted on EC2 which runs docker image. I personally go with FastAPI. - DynamoDB to store user attributes. - Make sure the backend is secure. Certs via Let's Encrypt. And there you have it folks, a complete guide to building a product.

Why Firebase is a great solution to build an MVP? Firebase is often considered a great solution for building Minimum Viable Products (MVPs) due to several reasons: 🔷Rapid Development 🔷No Server Management 🔷Realtime Database 🔷Authentication and Authorization 🔷Scalability 🔷Cost-Effective 🔷Easy Integration with Frontend Frameworks 🔷Comprehensive Ecosystem 🔷Analytics and Monitoring 🔷Community Support and Documentation #mobileapp #firebase #startup #smallbusiness #ceos #founders