I'm Indrajeet — a full-stack developer who likes turning questionable ideas into working products.
I'm especially interested in backend systems, AI, automation, APIs, and the messy technical problems that appear when an idea moves from “this would be cool” to “okay, now make it actually work.”
I usually start with a problem, an idea, or sometimes just a question that won't leave me alone.
Then I figure out what needs to exist.
If I don't know how to build something, I learn it.
Then I build it, break it, fix it, and eventually ship it.
The technology usually comes after the idea.
I like the parts users don't see — APIs, databases, authentication, queues, integrations, and the systems that make everything else work.
If something requires a human to do the same boring thing repeatedly, I immediately start wondering whether it should be automated.
I'm interested in using AI as part of an actual product rather than just putting a chatbot on a page.
Some ideas are useful. Some are weird. Most are both.
Started from a simple annoyance: repeatedly filling out startup applications. Turned it into a browser automation problem.
Started from an absurd idea and turned it into a real data-driven product with moderation, a database, and a working backend.
Started from the idea that an AI coding assistant should guide someone toward an answer rather than immediately giving it away.
Started from the question: what if technical interviews didn't need a human interviewer to be useful?
I don't want to be the developer who knows the most technologies.
I want to be the developer who can look at an unfamiliar problem and figure out how to build the thing.
That means learning new tools, understanding systems I haven't worked with before, and being comfortable not knowing the answer at the beginning.
I enjoy building side projects, experimenting with new technologies, breaking things and figuring out why, exploring weird product ideas, and learning by actually building.
Building things.
Learning things.
Probably automating something I shouldn't.
If you found something interesting here, check out the projects. That's where most of the interesting stuff happened.