When I started working as a Data Analyst at Synkron, it was my first time contributing to real business problems in a professional environment.
My role focused on working with structured datasets related to demand-driven supply chain systems.
I supported analysis around trends, variability, and data quality, and used statistical methods to help with forecasting and planning use cases.
At that stage, my work was primarily data-focused.
I interacted with existing machine learning models as tools rather than systems I fully understood.
Data went in, predictions came out.
I could evaluate outputs, compare metrics, and communicate results to stakeholders, but I didn’t yet understand how models were built, trained, or tuned.
This experience still mattered a lot.
It showed me how messy real-world data is, how constraints shape technical decisions, and how analytics fits into business workflows.
I learned how to structure problems, communicate insights clearly, and think about impact beyond just code or models.
Synkron became a bridge between theory and practice for me.
It helped me realize that building AI systems is not just about models, but about data pipelines, evaluation, communication, and making decisions in imperfect conditions.
That realization pushed me to move beyond using machine learning as a black box and start learning how these systems actually work — which later shaped how I approached my own software and ML projects.