My Data Analysis Journey: From Pascal to Python and AI
By: Roman Myskin - Nov. 1, 2025This year was tough, and there’s still so much to learn, but I decided to take a pause and review my educational journey in Data Analysis.
My path started back in high school, when the position of “data analyst” didn’t even exist. I’ve always been a fan of video games, and to create one, you need to know programming. I bought a book about Pascal and started exploring this blue treasure. I managed to create just one game - a flying rocket dodging asteroids. Most of my skills I practiced through math exercises, and I even participated in a few Olympiads.
In my adult life, I never thought that the skills I had in the Microsoft Office suite as a kid would be so useful. Just like in school, I realized these are must-have tools for everyone. Excel and Word are essential, but I’m still surprised when I meet people who can’t use VLOOKUP or PivotTables. And I especially can’t understand people who work with typing but haven’t mastered touch typing.
The next set of skills I gained was from Moscow State University, where I studied Physics - though honestly, I’ve forgotten most of them by now. I remember working with triple integrals and memorizing the elementary charge constant. We also studied statistics, higher mathematics, and linear algebra. The last one I especially regret forgetting, and next year I plan to master it again.
In 2019, my journey in Digital Marketing began. In the first six months, I worked hard to earn my Google Skillshop certificates. My skills in Google Sheets grew quickly as I learned shortcuts and built complex formulas. Eventually, I realized that advertising platforms don’t provide enough metrics to analyze data thoroughly and efficiently, so I decided to learn Power BI, Python, and a bit of JavaScript. The most valuable outcome of this experience was gaining marketing domain knowledge - which is just as important for a data analyst as knowing Python.
In 2025, my technical stack includes Python (pandas, NumPy, scikit-learn, SciPy, SQLite3, BeautifulSoup, matplotlib, seaborn); SQL (joins, window functions, CTEs, aggregations); Google Apps Script; Git; BigQuery; Azure; Looker Studio; Power BI (DAX, Power Query); Tableau; Google Analytics; Google Tag Manager (JavaScript); and Mixpanel. There are also some tools I’ve learned but haven’t yet had the opportunity to work with, such as MongoDB and Snowflake.
And I know there’s still so much to learn. The AI era has begun, and I sometimes feel like I’m falling behind - so next year, I plan to focus on TensorFlow, PyTorch, LLMs, MCP, and Hugging Face. I also have unfinished business with JavaScript, and this is something I don't want to rely on AI only.
Every stage of my journey has taught me one thing - learning never stops. Technology changes, tools evolve, and what once seemed advanced becomes basic in a few years.
I’m excited about what’s coming next - diving deeper into AI and bridging the gap between data analysis and intelligent automation. The best projects are still ahead.
Home