My mother never asked me what I wanted to study. She asked me what I wanted to build. It was not a distinction she had learned from a college brochure, it was the only framework available to someone who had watched systems fail and learned that the only reliable response is to make something yourself. We arrived in the United States in June 2024, after years of displacement from Afghanistan through the UAE, and the first thing I noticed was not the opportunity but the distance between where I stood and the institutions that held it. No one in my family had navigated a FAFSA, a transfer credit evaluation, or a conversation with an academic advisor. Every step required learning a language that was not just English but institutional, a dialect no one teaches but everyone assumes you speak. I learned it. I completed two associate's degrees simultaneously, published peer-reviewed research, and built software systems used by a Princeton nanofabrication lab and a nonprofit serving people in recovery, all while working and navigating an active immigration case. What that journey gave me was not toughness, a word people reach for when they want to admire difficulty without sitting with it, but clarity. I know exactly why I am studying computer science, because well-built systems reduce the distance between people and what they are trying to reach. My experiences did not inform my purpose so much as they became it.

The bug had been there for three days. Not a dramatic failure, just a quiet inconsistency, a number that was off by a fraction in a way that should not have mattered but did, because the system downstream trusted it completely. I had written the pipeline, tested it, deployed it, and still it was wrong. I stayed with it, tracing each function back to its assumption, until I found it: a single line that treated local time as UTC, a mistake so small it was almost philosophical. Fixing it took four seconds. Understanding why it had slipped through taught me more than the original build had. That is when I decided computer science was not a subject I was studying but a way of thinking I was developing, one that rewards patience, precision, and a genuine willingness to be wrong. Since then I have built automation tools for a nonprofit operations team, developed a file access system for a nanofabrication lab at Princeton, and published research on platforms that help students learn from simulations rather than memorization. Each project has deepened the same conviction: software built with care can quietly expand what people are able to do, and the engineering underneath it matters as much as the interface anyone sees. My goal after graduation is to work as a backend or DevOps engineer, building the infrastructure that other systems depend on, the layer that has to be correct even when no one is watching it. From there I want to move into MLOps, where I can help organizations deploy machine learning responsibly, closing the gap between what a model can do in a notebook and what it can actually do in the world. I am a first-generation student who has worked, built, and published alongside completing my degree, and I will carry this investment into engineering that tries to be worth the trust placed in it.

My name is Abubaker Ahmadi, though most people know me as Saba. I began my education in Afghanistan, lost it for several years when conflict forced my family through the UAE, and rebuilt it from scratch after arriving in the United States in June 2024. That gap taught me something I have never been able to put down: education is not a building or a credential, it is the daily decision to understand something you did not understand yesterday. When school was taken away, the wanting to learn was not, and that distinction has shaped every choice I have made since. At Mercer County Community College I earned two associate degrees with a 4.0 GPA, founded a campus computer science club that grew from thirteen members to sixty-four, and took an internship at a Princeton research lab where I built an open-source system that gives researchers reliable access to their work. This fall I transfer to Rutgers University to finish a degree in computer science and mathematics. I chose those fields deliberately. Mathematics was the first language that made sense to me. Computer science is how I extend a teacher's reach beyond a single room, by building systems that keep working after the person who built them has gone. My goal is to become a software engineer who builds the infrastructure other people rely on without ever seeing it, the layer underneath the applications, and eventually the systems that let machine learning run reliably for the people who depend on it. Longer term I plan to pursue a master's degree and keep widening access to the tools that let people learn and build. I know firsthand what it costs when that access depends on where you happen to be standing when circumstances change. I want my work to lower that barrier for the next person. This scholarship would let me keep choosing the long path. I live at home and commute to keep costs down, and I have accepted only the financial aid I can responsibly repay, which means there is constant pressure to take the nearest paycheck instead of the harder, more valuable project. Financial support loosens that pressure. It would protect my time for the deep skill-building and the unpaid, high-growth work that turn a self-taught student into someone who builds things that last. I have already proven I will do the work without anyone requiring it. This support would let me do more of it.

Sine θ equals the opposite over the hypotenuse. On paper it is a ratio, memorized and then left behind. For me it is a tent in Kabul, a carpet thin enough to feel the earth through it, a textbook passed between hands until its spine cracked open. School in our village was never promised, only assembled when enough pieces aligned. I was thirteen the afternoon our principal called my name between lessons. A teacher was leaving, he said, and he had heard I had a gift for mathematics. Then he walked me to a tent where thirty students sat shoulder to shoulder, waiting for a teacher who was not coming. Their eyes did not just look at me, they leaned toward me, as if hope itself carried weight. Standing there I asked a question that has never left me: what pushes a child to wake before dawn and walk through active conflict to sit in a tent taught by another child making it up as he goes? My mother, who never attended school, would have understood them instantly. When you cannot learn, she would say, you can only survive what happens to you, and when you can, you shape what happens next. I turned to the board and wrote sin θ, and I understood what I was meant to do. The years after did everything they could to take that away. Conflict scattered my education across three countries, and for a long stretch there was no school at all, only motion and waiting. When I reached the United States in 2024 I had no map and no one to tell me which door to knock on. So I did what that tent had taught me. I taught myself, the slow way, the way that holds. At Mercer County Community College that habit carried me to a 4.0 across two associate degrees and into a research lab building software, and this fall it carries me to Rutgers to finish my degree in computer science and mathematics. I chose those fields on purpose. Mathematics was the language that tent first handed me. Computer science is how I keep teaching when I cannot be in the room, by building systems that persist after the teacher vanishes. I want a career building the infrastructure people rely on without seeing it, and the work I have already started points there. An open-source file system for a research lab, automations that hand hours back to overworked staff, all of it came from the same instinct that tent gave me: lower the barrier, make the tool reach further, let the person on the other side get to what matters. This scholarship would let me keep choosing the long path. I live at home, I commute, and I have accepted only the aid I can responsibly carry, which means every semester quietly pulls me toward the next paycheck and away from the next idea. Financial support loosens that pull, protecting my time for the harder projects and the graduate study that turn a self-taught student into someone who builds things that last. Sin θ. A ratio I first wrote for thirty students in Kabul. I am still solving for it.

Hands-on chip fabrication is one of the hardest skills in technology to teach. The equipment is cost-prohibitive for schools, and the facilities that house it, semiconductor fabs, cost upwards of $15 billion each to build. Students are effectively locked out of the field that underpins modern AI infrastructure, and the industry faces a widening talent shortage as a result. To address this, I led the R&D of a Digital Twin simulation in collaboration with Princeton University. The goal was to move training for the Advanced Device Packaging process out of the physical lab and into a 3D interactive virtual environment that matches the real world in both visual form and mechanical function. Over six months, mentored by Bert Harrop, a staff researcher at Princeton, my team of six studied the full chip packaging lifecycle to make the simulation procedurally accurate. We reconstructed every machine, tool, and lab detail as high-fidelity 3D models, imported them into a game engine, and programmed the mechanical logic so that each virtual machine moved with the same speed and axis limitations as its physical twin. To turn the model into a usable training tool, we built a VR framework for immersive interaction on headsets such as the Meta Quest. Users perform the exact task sequence required in the physical lab rather than passively observing it. To replicate expert oversight, we integrated two AI agents: an Instructor with spatial awareness that guides users verbally through each step, and an Evaluator that assesses competency by quizzing the student. The result compressed a multi-billion-dollar facility into a downloadable file. In June 2025, I was selected to present our work at the TechConnect World International Conference in Austin, Texas, putting the project in front of the industry leaders dealing with the talent shortage firsthand. There is still real room to improve. Our desktop version is more accessible than VR but offers a weaker sense of physical interaction, so I plan to refine the desktop interface to better approximate the tactile experience for students who cannot afford headsets. The approach also generalizes beyond semiconductors, since any field that depends on human and equipment interaction can be simulated the same way. My longer-term goal is to scale this from a single university lab project into a national library of training simulations. This is what draws me to working in technology. A well-built system can take something locked behind billions of dollars in capital and make it available to anyone with a laptop. Software lets a small team remove a barrier that would otherwise keep thousands of students out of a field, and it lets the next generation of engineers learn through practice rather than observation. Building tools that widen access instead of narrowing it is the work I want to keep doing.