Learning Cloud by Building: My Journey from Python Backend Development to AWS
As a final-year CSE (AI/ML) student, I’m building my software engineering foundation through Python backend development, real projects, and AWS. This article shares why I’m choosing project-based cloud learning and what I plan to build along the way.
I’m currently in the final year of my B.Tech in Computer Science Engineering with an AI/ML specialization. Over the last few months, most of my attention has been on Python, backend development, SQL, Git, Linux, DSA, and computer science fundamentals.
One thing I’ve started realizing is that learning technologies separately is not enough for me.
I can read about APIs, databases, Docker, or cloud services, but unless I actually use them together in a project, I don’t really understand where each one fits.
So I’m trying to change the way I learn.
Instead of finishing one long course after another, I want to learn something, use it in a project, run into problems, fix those problems, and then continue.
What I’m building right now
My main project at the moment is called RiskOps.
It is an independent backend project based around a risk-investigation and case-management workflow.
The basic flow I’m working with is:
Entity → Risk Event → Alert → Investigation Case
The idea is to gradually build features around case creation, investigation workflows, status changes, audit history, users, permissions, and data storage.
My backend direction is mainly:
- Python
- FastAPI
- PostgreSQL
- REST APIs
- Git and GitHub
- Linux
- testing
- Docker
I’m still learning several of these while building, so I don’t consider RiskOps a finished product yet.
That is actually the point of the project.
I want it to become the place where I learn how different software-engineering pieces connect instead of building separate tutorial projects for every technology.
Where AWS fits into this
Until now, a lot of my development work has been local.
The application runs on my machine, the database is local, and I control the whole environment.
But real software obviously doesn’t stop there.
Once an application has to run for actual users, a lot of new questions appear.
Where does the backend run?
Where should the database be hosted?
How are secrets stored?
How are logs collected?
What happens if the application crashes?
How do permissions work?
How much does the infrastructure cost?
How do you know whether resources are being wasted?
Those are the questions that made me interested in AWS.
I don’t want to learn AWS by memorizing a long list of services. I want to understand the services when I actually need them.
For example, when I eventually deploy a backend, I want to understand why I’m choosing one compute option over another rather than simply following a tutorial.
The same applies to storage, databases, IAM, networking, monitoring, and security.
Cost is something I had not thought about enough
Another area I’ve recently become more interested in is cloud cost.
As a student, it is easy to think about cloud mainly in terms of deployment and scalability.
But infrastructure decisions also have a financial side.
The type of resource you select, how it is configured, how many instances are running, and how resources are purchased can all affect the final bill.
That made me curious about FinOps and cloud-cost optimization as well.
It is interesting because it sits somewhere between software engineering, infrastructure, data, and business decisions.
As I learn AWS, I want to understand not only how to deploy something, but also how to avoid building unnecessarily expensive infrastructure.
AI will probably become another layer
My degree specialization is AI/ML, and I also want to explore applied AI engineering later.
For RiskOps, I’m considering an AI layer for things such as:
- case summarization
- contextual retrieval
- investigation assistance
- risk insights
- decision support
But I don’t want to add AI just because it sounds impressive.
My priority is to first make the underlying software work properly.
If I add an AI feature, I want to understand what problem it solves, how its output can be checked, and what happens when the model gives a wrong answer.
How I want to use AWS Builder Center
I’m using AWS Builder Center as part of this learning process.
My goal is to strengthen my cloud fundamentals, work through hands-on material, and eventually connect what I learn directly to my projects.
I’m also planning to work toward an AWS certification, but I don’t want the certificate itself to become the goal.
For me, the better outcome would be reaching a point where I can look at one of my own applications and explain:
- how I would deploy it,
- how I would secure it,
- how I would monitor it,
- how I would store its data,
- and how I would control its cost.
That would mean I actually learned something useful.
Where I want this to lead
My current direction is becoming clearer:
Python Backend → Databases → Cloud → Applied AI
I still have a lot to learn, but I would rather document the process while I’m learning than pretend I already know everything.
My next goal is to gradually take RiskOps from a local backend project toward something that can actually be deployed and operated in a cloud environment.
I’ll use future posts to document what works, what breaks, and what I learn along the way.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article