Machine Learning Essay
Layli Nurul Fatimah Zahra
What first drew me into the world of Machine Learning (ML) was something as usual
as scrolling through Instagram during high school. I noticed how the app’s algorithm seemed
to "know" exactly what I wanted to see whether it was memes, travel photos, or anything I
have watched. At first, I found it almost scary, as if the app could read my mind. Over time, I
realized this "mind-reading" magic was actually a sophisticated recommendation system
analysing my behaviour: the posts I liked, my search history, even how long I lingered on
certain content. This revelation sparked my curiosity like how could an app process vast
amounts of data so quickly and deliver hyper-personalized results in real time?
This question encourage me to explore how the mathematical concepts I was learning
could explain this phenomenon. Through my linear algebra and statistics courses, I began to
see how matrices and probability distributions could represent user preferences and content
relationships. I was fascinated by how simple mathematical operations - matrix
multiplication, vector spaces, and probability calculations - could turn raw numbers into
intelligent predictions. The more I learned, the more I wanted to understand not just the
“what” of these systems, but also the “how” of the math behind them. Calculus and discrete
mathematics courses in particular helped me understand the optimization principles that make
personalized recommendations possible.
To turn my curiosity into knowledge, I took advantage of free courses offered by my
university, which introduced me to data visualization and ML fundamentals. Through these
classes, I learned to use tools like Google Colab for running Python code and libraries
like Pandas and Matplotlib to clean and visualize datasets. Though the courses were
introductory, they gave me the confidence to explore further on my own.
I started experimenting with small projects, often using datasets from Kaggle or public
repositories. One of my early projects was analysing weather data from my city to predict the
next day's air quality, humidity levels, and weather conditions. Although the accuracy of my
model was far from perfect (mainly due to data limitations and challenges in feature
engineering), this process taught me valuable lessons. I learned the importance of data
preprocessing, the impact of feature selection on model performance, and the need for
rigorous validation techniques. These projects, though simple, reinforced my interest in ML. I
realized that I didn't just want to use applications powered by ML-I wanted to build them and
contribute to their evolution.
My dream is to work on modern Machine Learning systems, whether in social media
(like improving those Instagram algorithms that fascinated me) or in fields like start-up or
finance. However, I know I need structured training to master the technical and operational
complexities of MLOps. That is why I need SISTECH’s portfolio program. I’m particularly
excited to make Hands-on MLOps Training, making projects, and mentorship by amazing
mentor. With SISTECH’s guidance, I aim to evolve from a self-taught enthusiast into a skilled
person capable of delivering ML solutions that are not just innovative but also strong,
scalable, and ethical.
My journey into Machine Learning began with a sense of wonder about everyday
technology. Now, it’s a deliberate pursuit of expertise one that balances curiosity with rigor.
I’m eager to contribute to the future of Machine Learning, and I believe SISTECH’s program
is the perfect catalyst to turn my aspirations into impact.