Huh? Am I breaking the rules or something? I don't know what you are saying sorry but it isn't useful to this discussion.
From the FAQ:
What to Submit
On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity.
I asked others how they would build a chat system in 3 days, and my intention is to use that to understand how to solve my own question.
I used to care about that too until I got enough money (not a TON, but enough to put my worries aside for 5-10 years) from equity working at a smaller company.
Now I'd rather keep a simple job and work on side projects that interest me than working at a FAANG.
Petfood Express. It’s a California pet store chain with like 40 stores or something but the staff there are so nice and helpful. Their website is also nice to use though I much prefer going to the store.
The staff is knowledgeable and pet lovers first. I wanted to give my cat the best food and they were so helpful in helping me pick a few different items to try. My cat loved them all so that’s a great success!
I was in the same boat as you as a mostly self taught engineer. What really helped me personally are the following and I would like to share:
* Read “Clean Code” and try to apply the principles even in toy projects. I only have five years of experience but this is the most profound book so far in my career, and it’s one that is good to re-read every so often to keep it fresh and apply :)
* If you’re at a mid size company now with other engineers judging your work, then this is a great opportunity to review their PRs and see how they’re going about their business. You have a decade of experience and I’m sure you’ll both have suggestions and learn something. Hop on a video call and pair with someone senior or staff at your company.
* Do you have any ideas that you wish you could bring to life? Even a small ToDo app or something like that. If you can plan the requirements, design the application architecture, and choose a good set of libraries and tools to aid you, you will feel a huge surge of confidence that you have the ability to create. This is important. This would also aid you if you decide to become a PM because if you’re in charge of making a vision come true, what better than to be technically aware of what’s really going to make that vision a reality.
Just a couple things that helped me. All the best, and remember we are both student and teacher til our last breath so don’t sweat it if someone is younger/older at different stations in life.
Long post, but I hope it helps someone out since it covers web,mobile, etc. MVP to me means being able to build something quick and iterate upon, so while it could be a throwaway project each time it still makes sense to have a preferred solution for each problem space.
I stick to tech I know best and also languages which offers a vast variety of available SDKs and dependencies to consume so I don't need to waste time re-inventing the wheel. I also try to plan for my requirements up front so I can best prepare. In this case, I am building an application that requires authentication, notifications, private + group chat, location awareness, and an admin sites for the CMS, partner portal, etc.
- For data storage, I use postgres for persistence, and redis for caching and other temp things. I'm not an expert at either of these, but I have used both extensively and have enough experience/interest to work through any issues that may come up.
- I use nodeJS + express to write backend services. Example: My web and mobile clients consume a REST service that interacts internally with a data layer service. Since the code is organized very simply and modular, it makes it easy for me to spin up new services that consume existing services, or makes it easy to extend. I have a real time service that could use the same data layer service if needed.
- I try to use SDKs that can solve my problem cheap + quick. This is where planning up front is also valuable because if you can identify which SDKs are available for the languages you are using, you will save lots of time and pain. For example, I'm using connectycube (free up to 10k users) for video + chat because they have flutter SDKs it takes literally 10 minutes to set up and be most of the way there. I use firebase for authentication (its free) on the clients, and for my REST service token verification. Firebase can also handle push notifications, so keep that in mind! I use mapbox for geocoding because it is cheaper than google maps.
- For infrastructure, I was using AWS but switched to DO managed postgres and droplets. I'm still using S3 and cloudfront.
The reason I did this is because I don't have much infrastructure experience and made a few misconfiguration mistakes on AWS that bit me. DO makes it dead simple for now so I can focus on just building my product and maybe some day I will be lucky enough to work with a team who knows AWS :)
- For web, I have the website (hosted using carrd, dead simple website builder. They have a cheap paid plan that lets me do things like add a contact us widget and add a google analytics), while the CMS and partner sites are being built using create-react-app. There is no consumer web app (mobile app only), so really those users will just see the site and go download the apps. Partners can access the partner portal to do things like adding their business listing and any promotions, etc.
- Mobile apps are built using flutter. I've dabbled with swift/swiftUI, kotlin, react-native, and flutter and today I prefer flutter for complex projects that I need on both iOS and android. Regardless of what path you choose, some tips I can offer when you're in the middle ground between a "this is more complicated than what's in the /examples folder" and "any top app that you use on a regular basis":
• Spend a weekend going through the platform specific design guidelines. Open up your favorite top apps on iOS and android and use the same features on both. Notice the subtle differences.
• Read the app store guidelines for both so you know what to expect.
• Plan your application structure. What are the shared states, and how will you share data? Do you need to persist data on the device? What device permissions do you need to manage? What are the top level views? Do you need to make network calls? I used to never do this because I figured hey, its just a toy project!! But, as soon as that toy project turns into a passion pit, I need to make that new features can be added very fast instead of requiring a re-write. Just note the different moving pieces, and organize your code in a way that makes sense.
• Nothing is perfect - flutter web is kinda meh and honestly I think its best anyway to just have separate mobile code, and web code.
Learned a ton over the pandemic, hoping to put all of this to good use and keep learning. All the best!
I’m an American so I always thought it’s cool you have NHS because we have to either pay for our own health insurance or we get it from our employer. What do you mean by health insurance there - are those additional offerings that NHS doesn’t provide?
From the FAQ:
What to Submit On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity.
I asked others how they would build a chat system in 3 days, and my intention is to use that to understand how to solve my own question.