Academic Preparation

During my undergraduate study in Information Management at Zhejiang University of Finance and Economics, I became interested in how data systems can turn operational signals into decisions that are both timely and explainable. My cumulative GPA was 3.62/4.00, with stronger performance in database systems, statistics, and software engineering. More important than the transcript alone, however, was the habit of tracing a result back to the data definition, processing step, and evaluation measure that produced it.

In a capstone project on retail demand forecasting, I worked with a four-person team to clean two years of daily transaction records, design features for promotions and holidays, and compare gradient boosting with a seasonal baseline. I owned the validation pipeline and discovered that a random split overstated performance because it allowed future promotion patterns to leak into the training set. After replacing it with a time-based split, the mean absolute percentage error increased from 11.4% to 15.8%, but the result became operationally credible. This experience taught me that reliable computing begins with disciplined problem framing.

Evidence of Readiness

As a data product intern at 杭州知衡科技有限公司数据产品实习, I translated requests from operations managers into data definitions, SQL checks, and dashboard acceptance criteria. When different teams used “active customer” to mean different things, I proposed a shared metric note and a small set of reconciliation queries. The task was modest, but it made visible the communication layer between technical implementation and organizational use. I want graduate study to help me deepen that layer through stronger foundations in algorithms, systems, and applied machine learning.

Why This Programme

I am applying to the Master of Computing (General Track) at National University of Singapore because its application framework asks candidates to connect motivation, preparation, and future plans rather than present a list of disconnected achievements. The programme’s computing environment is a suitable next step for my goal of becoming an engineer who can build dependable data products from problem definition through deployment and evaluation.

I am especially prepared to contribute a practitioner’s perspective to project-based learning: I have seen how a technically elegant model can fail when its data contract is unclear, its evaluation period is inappropriate, or its output cannot be explained to a non-technical user. I also understand that programme and faculty details must be checked against the current official application page before submission; I will not claim a specific module, supervisor, or research facility unless it is confirmed for the intake to which I apply.

Career Direction

After graduation, my immediate goal is to work as a data or machine-learning engineer in a Singapore or Asia-Pacific technology team, focusing on trustworthy forecasting and decision-support systems. Over the following three to five years, I hope to lead projects that connect model quality with measurable operational outcomes, including data lineage, monitoring, and responsible communication of uncertainty. The long-term direction is not a promise of a particular job or outcome; it is a concrete learning agenda grounded in the problems I have already encountered.

To prepare for this path, I will bring the discipline learned from 完成需求预测看板并将周度人工整理时间缩短约30% into graduate work: define the user and decision first, make assumptions explicit, test against realistic time periods, and document limitations. I believe this combination of technical depth and practical accountability will allow me to learn seriously and contribute constructively to the programme community.