DEV Community: Raiyan Memon

Coding Is Getting Easier. But System Design Still Matters.

Raiyan Memon — Sat, 22 Aug 2026 19:55:06 +0000

Writing code is becoming easier.

With AI tools, you can generate APIs, database queries, UI components and even complete features much faster than before.

But building software isn't only about writing code.

The bigger challenge is often how the entire system should work.

Coding a Feature vs Designing a System

Take a URL shortener.

Writing the basic version isn't very difficult:

Client
↓
API
↓
Database

It works.

But what happens when you have millions of requests?

Now you need to think about caching, database bottlenecks, queues, availability, failures, traffic spikes and capacity.

The code might still be simple.

The architecture isn't.

That's where system design becomes important.

The Problem I Had

While learning system design, I noticed that most of the learning was passive.

I would watch someone design a system, understand their decisions and think:

«"Yeah, that makes sense."»

But when I got a blank page and had to design the same system myself, it was much harder.

That's when I thought:

Why isn't there a simple way to actually practice system design instead of just watching it?

So I started building one.

Introducing Devscape

Devscape lets you take a system-design problem and build the architecture yourself.

You get a blank canvas.

You add your services, databases, caches and connections.

Then you submit your design.

Instead of simply showing you a model answer, Devscape evaluates your architecture using a rubric engine.

For example:

Caching ✓
Scalability ⚠
Failure Handling ✗
Capacity Planning ⚠

The interesting part is that I'm not trying to make the grading:

«"You didn't add Redis, so you lost 10 points."»

There isn't always one correct architecture.

A simple architecture can be perfectly valid if it satisfies the requirements.

The goal is to evaluate whether the architecture makes sense for the given requirements and constraints.

You can try it here:

https://devscape.in

Why I Think This Matters

AI can help us write code much faster.

But someone still needs to decide:

How the system should be structured
Where the bottlenecks will be
How it should handle failures
How it should scale
Which trade-offs are worth making

As writing code becomes easier, I think these decisions become even more important.

That's what I'm trying to explore with Devscape.

It's still early, and I'm actively improving the rubric engine.

If you try it, I'd genuinely appreciate your feedback — especially if you think the grading is wrong.

https://devscape.in

Have recommendations or thoughts? Feel free to share them in the comments!

Want to know more about me or get in touch? Visit my website

Keep coding and exploring ❤️

Determine Server Capacity to handle the requests

Raiyan Memon — Sat, 19 Oct 2024 09:53:38 +0000

Knowing how much load your server can take at once is crucial.

In this blog, we will learn how many requests your server can handle at once.

We are going to fire an Apache Command with some configurations in it to know the capacity of our server and push it to its limit so that we get to know whether to increase the size of our server or add a load balancer to it.

A command line tool that we are going to use is known as ApacheBench (ab). It tests the performance of a server by sending a specified number of requests and measuring the response times.

Also provides insights like requests per second, time taken per request, and transfer rates.

Copy and paste the below command on the local or live server works but make sure the apache2 server is installed.

ab -n 1000 -c 10 https://raiyanmemon.in/

If you are getting any error try with “http” rather than using “https”.

It mostly throws an error of ssl with “https” just on the local server but works same.

Here’s a breakdown of the parameters:

ab: Apache Bench.
-n 1000: The total number of requests to send. In this case, 1000 requests.
-c 10: The concurrency level, meaning 10 requests will be sent simultaneously.
https://raiyanmemon.in/: The URL of the server you are testing. At last, the forward slash (/) is mandatory.
You can change the number of requests or the concurrency level if you want to create a heavy load on your server.

You can also change your server URL to an IP address something like this: http://192.168.1.17/

Your Response to the above command will be like this:

Server Software:        *******
Server Hostname:        raiyanmemon.in
Server Port:            443
SSL/TLS Protocol:       *************
Server Temp Key:        ********
TLS Server Name:        raiyanmemon.in

Document Path:          /
Document Length:        9607 bytes

Concurrency Level:      10
Time taken for tests:   4.202 seconds
Complete requests:      100
Failed requests:        0
Total transferred:      1029945 bytes
HTML transferred:       960700 bytes
Requests per second:    23.80 [#/sec] (mean)
Time per request:       420.214 [ms] (mean)
Time per request:       42.021 [ms] (mean, across all concurrent requests)
Transfer rate:          239.36 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:      119  136  48.9    123     344
Processing:    40  129 259.1     42    1317
Waiting:       40  128 259.1     42    1316
Total:        160  264 302.3    165    1661

Note: Just spend some time to analyse the response and you will get to understand how simple it is.

After hitting the command to know whether the command has an effect or not, just log in to your server and see the CPU usage you will notice an instant hike.

If you didn’t notice any change on the CPU try increasing your number of requests and concurrency level.

This is it. You are ready to test your server capacity to handle the requests.

Feel free to comment if you have any recommendations.

Keep coding and exploring ❤️

Running Laravel Queue Worker in Production

Raiyan Memon — Thu, 17 Oct 2024 17:25:55 +0000

Laravel Queue Worker plays an important role in running in the background. It does not put a load on our web application.

For firing mail and any other stuff that takes time for such tasks, we can schedule a queue worker.

In this we are going to see how we can schedule our queue worker on the production server.

Most people including me had a hard time running this command on the production server we can just keep running “php artisan queue:work” every time so I got the solution.

Go inside your code folder on the server and just place this command and hit enter twice.

nohup php artisan queue:work --daemon &

Congratulations! You just fired a command which will keep the worker running even if you close the console of your server.

One more thing if you have made any changes in the queue worker code and something related to that and the changes do not reflect on the production link just restart the queue with this command.

php artisan queue:restart

This will stop the command of keeping up the queue worker and then run

php artisan optimize:clear

This command is mandatory because Laravel stores the cache and does not clear it by default.

After this just hit the first command and your queue worker is automated again.

Keep coding and exploring❤️