The article can follow one photo from the user’s laptop all the way through a cloud architecture.
Start with:
You select a photo.
Click Upload.
A progress bar moves from 0% to 100%.
Done.
Simple, right?
Not quite.
In those few seconds, your file may have been authenticated, validated, encrypted, transferred across the internet, stored across cloud infrastructure, scanned or processed, and prepared for delivery to users around the world.
So what actually happened after you clicked Upload?
Then we follow the journey.
1. Your application first decides whether you’re allowed to upload it.
Authentication and authorization come first. Who is this user? Can they upload files? What type? How large?
This connects nicely to the API-security article without repeating it.
2. The file begins its journey to cloud storage.
Introduce Amazon S3 and explain object storage simply. Rather than thinking of it like saving something to C:\Photos, the application stores an object inside an S3 bucket with a unique key.
We can also explain presigned URLs, because that’s a genuinely useful technical concept.
Instead of:
Browser → Backend → S3
for large uploads, an architecture could allow:
Browser → S3
using a temporary presigned URL generated by the backend.
The backend controls permission without becoming the middleman for every byte of the file.
3. S3 receives the object—but we’re not necessarily finished.
Now we introduce event-driven architecture.
The new S3 object can trigger an event.
For example:
Photo uploaded → S3 event → Lambda
Lambda might resize the image, generate a thumbnail, extract metadata, or kick off another processing workflow.
For video, the processing could be much more involved.
So the user clicked one button, but several systems can begin working behind the scenes.
4. What if the upload is huge?
Great place to explain multipart uploads.
Instead of trying to send a 5 GB file as one enormous request, it can be divided into multiple parts and uploaded separately before being assembled.
If one part fails, you don’t necessarily need to start the entire upload again.
5. How does someone download/view it later?
This brings in Amazon CloudFront.
Instead of every user around the world repeatedly retrieving content directly from its origin, a CDN can cache and deliver content closer to users.
Now we can explain the difference between storing something and delivering it.
S3 stores the object.
CloudFront can help deliver it efficiently.
6. Is my file just public now? 👀
Very important section.
No.
Explain bucket permissions, IAM, encryption, presigned URLs, access controls, and why making an entire S3 bucket public just because your website needs images is usually not the architecture you want.
Then bring the entire journey together:
Click Upload
↓
Authenticate
↓
Validate
↓
Generate upload permission/presigned URL
↓
Upload to S3
↓
Encrypt & Store
↓
Trigger processing
↓
Lambda processes file
↓
CloudFront delivers content
↓
The user sees their photo
And the bigger lesson becomes:
The cloud isn’t simply “somewhere your files go.”
It’s an architecture of storage, security, networking, processing and delivery working together to make that Upload button feel boringly simple.
Then Arthurite:
“At Arthurite Integrated, we design cloud systems behind experiences like this.”
Storage architectures, S3, serverless processing, CDN delivery, security, backups, APIs, etc.