Startup: What, Why, and How?
In my first college semester, a classmate and I discussed how often we asked our parents for money. We wanted a small source of income. We chose web work, although neither of us knew web development.
We found our first client at ushodayanetworks.com. The first site was a bridal collection gallery. I first tried to copy code from the browser tools on kalkifashion.com. It did not work. We then tried another tool that produced a starting codebase. That result was not good enough either.
After some research, we found v0. We used it to make kavyareddy.in before we understood web development. We had trouble with deployment and domain setup, but we launched the site.
That work taught me an important lesson: learn the system instead of acting as if you know it. I began to learn JavaScript. At the same time, I built ushodayanetworks.com with React.
Client work soon became more than a way to earn money. It made me learn. Each project gave me a real problem with a real deadline. That experience later gave me the confidence to work on Anokha 2026 and build On Duty Amrita.
I learned that software is not only visual design. It is also logic, data flow, and system design. That led me to data structures and other engineering topics.
My View of AI Tools
Starting with v0 was useful. It helped me make a prototype quickly. It did not replace the need to learn.
AI can help you make a prototype. It cannot build your career for you.
When kavyareddy.in had deployment problems, I could not solve them until I understood DNS records. That taught me another rule.
Use AI to work faster on things you understand. Do not use it to hide gaps in your knowledge.
Where We Are Now
The goal has changed. I no longer build only to avoid asking my parents for money. I build because I enjoy solving problems.
Ushodaya Networks was the place where I changed from a student who copied code into a person who could build systems. We moved from simple sites to custom RFID work and backend experiments.
If you want to start, start before you feel ready. Take the work. Make mistakes. Use AI tools. Then read the documentation and learn why each mistake happened.
The Student Part of the Double Life
My GitHub may suggest that I do very well in every computer science class. I do not. My grades are often average, and I accept that.
I must divide my time between learning, clients, projects, and exams. Some exams focus on topics that do not feel useful in their current form. That makes study difficult for me.
The Builder's Problem
I can work for many hours on a real problem, such as server latency or a chess PGN analyzer. I find it harder to memorize textbook definitions without a clear use.
I still understand why older methods appear in a course. They show why newer methods exist. But I value one skill most: the ability to solve a new problem.
I do not treat college as a race for a CGPA. I treat it as a resource. When a subject helps me build better software, such as data structures or operating systems, I study it deeply. For other subjects, I do enough to pass.
The Balance
This double life is not a plan to do both things perfectly. It is a daily tradeoff. Sometimes I accept a lower grade to ship a product. Sometimes I pause a project to protect my semester.
It is stressful and imperfect. Still, the skills I gain from real work are the skills on which I am building my future.
