I Build Software That Thinks, Connects, and Scales

From Building Features to Building Systems

I’ve always been more interested in what happens behind the feature.

A button is easy to build. What interests me is everything that happens after it is pressed: how the request moves through a system, how data is retrieved, how different services communicate, how access is controlled, and increasingly, how AI can reason within that workflow and take meaningful action.

That curiosity has shaped the way I approach software engineering. I don’t want to simply make applications that work. I want to understand why they work, how they behave under complexity, and how they can become better systems over time.

Where I Am Today

I’m a computer science graduate-in-progress turned software engineer, currently focused on agentic AI and backend engineering.

My work sits at the intersection of traditional software engineering and intelligent systems. I build with TypeScript and JavaScript across the stack while going deeper into backend architecture, databases, distributed workflows, LLM orchestration, retrieval, state management, and AI-powered applications.

My current focus is OrchestrAI, a multi-agent customer-support system I’m engineering around the idea that an AI application should be more than a chatbot.

Instead of treating an LLM as a single box that receives a prompt and returns an answer, I’m exploring how routing, retrieval, identity verification, conversation state, deterministic tools, and multiple specialized agents can work together as one controlled workflow.

Building it has pushed me to think about AI from an engineering perspective: state, reliability, boundaries, data, orchestration, and what happens when things don’t go as planned.

Before AI, I Learned by Building

My journey didn’t begin with agentic systems.

One of my most formative projects was CampusNexus, a full-stack campus platform built around a simple problem: too many parts of student life were scattered across disconnected channels.

I turned that idea into a connected system for societies, events, study groups, mentorship, chat, notifications, and administration. What started as a university project became an opportunity to work through real engineering problems: backend architecture, authentication, role-based access, real-time communication, data relationships, and a growing API surface.

That experience taught me an important lesson:

A good application isn’t a collection of features. It’s a system where those features make sense together.

That idea still influences how I build today.

I Learn by Going Deeper

My projects have gradually moved from building interfaces and APIs toward understanding the systems underneath them.

I’ve worked with full-stack applications, real-time communication, databases, authentication, caching, Dockerized environments, and AI integrations. More recently, I’ve been spending more time on the backend and on the engineering problems surrounding AI from retrieval and embeddings to stateful agent workflows and deterministic tool execution.

I’m particularly drawn to problems where there isn’t a single obvious implementation.

The kind where you have to ask:

Where should this state live? What happens if this step fails? Should this action be deterministic or delegated to an agent? How does the system know what it is allowed to do? How do we make an intelligent system predictable enough to trust?

Those are the questions I want to keep getting better at answering.

More Than Code

Software has also taught me that engineering rarely happens in isolation.

Through university communities and leadership roles, including my time as a Microsoft Learn Student Ambassador Co-Lead, I’ve had opportunities to organize technical activities, collaborate with others, communicate ideas, and help build communities around technology.

I’ve also competed in hackathons and programming competitions, including two runner-up finishes, experiences that taught me to move quickly from an idea to something tangible while working within real constraints.

These experiences shaped another part of how I work: I enjoy building, explaining, collaborating, and figuring things out when the path isn’t already clear.

What I’m Building Toward

I’m still early in my career, and I see that as an advantage.

There is a lot I haven’t mastered yet and a lot I want to learn.

My direction is increasingly clear, though: AI engineering, backend systems, and the infrastructure behind intelligent software.

I want to go beyond simply integrating AI APIs. I want to understand how intelligent systems are designed, orchestrated, optimized, evaluated, and eventually operated reliably at scale.

For now, that means building projects, breaking things, reading deeply, experimenting, and turning every difficult problem into a better understanding of the systems behind it.

The goal is simple:

Build software that isn’t just intelligent but thoughtfully engineered.