For a long time, my development life existed within the predictable world of my local machine. I wrote code, it ran, and that was the extent of my world.
Few Months ago, I had chance to step outside of my comfort zone and dive into the world of Open Source. If I had to describe the feeling of that first moment, I would point to a specific scene from the Disney movie, "Ralph Breaks the Internet".
_
picture of movie "Rack it Ralph"
first time Ralph and Vanellope walks into the world of internet
Just like Ralph and Vanellope stood on that balcony, gazing wide-eyed at the endless, futuristic skyline, I felt completely small. In the movie, the Internet is described as a sprawling, infinite metropolis—bustling with flying vehicles, and towering skyscrapers representing the giants of the web.
Coming from the quiet, controlled environment of my local machine, the Open Source ecosystem felt like that futuristic city. The towering buildings weren't Amazon or Google, but massive repositories with millions of lines of code. The flying cars weren't just traffic. They were the large stream of Pull Requests, Issues, and Discussions happening in real-time across the world. People were continuously building, rebuilding, breaking and fixing the projects.
It was terrifying, yes. But just like Ralph looking out at that horizon, I realized the potential of this limitless world.
My Contribution Highlights
Driven by this excitement, I didn't want to just be a tourist in this new city. It was intimidating, but I am incredibly proud to say that I have successfully contributed to some of the foundational pillars of the Python data ecosystem.
I have had PRs merged into:
- Scikit-learn
- NumPy
- Pandas
- Dagster
Seeing my code become part of tools that millions of developers rely on was a exciting experience.
Why I Fell for Dagster
This realization explains why I fell so deeply for Dagster.
While exploring it, I got amazed by their core philosophy of Software-defined Assets. The concept of treating data not just as a byproduct, but as a first-class asset was very interesting. Treating data as assets shifts the focus from managing execution tasks to maintaining the freshness of the actual data products. This approach automatically generates clear lineage graphs, allowing you to easily understand dependencies and track how data flows through the system. As a result, debugging and collaboration become significantly more efficient because you are interacting with defined data outcomes rather than abstract code logic.
Reading the Dagster source code didn't feel like studying. I found myself mentally visualizing the entire process like how the data flows, how the assets are materialized, and how the engine handles dependencies. Simulating these complex data journeys in my head was incredibly fun and engaging.
Stepping out of my local machine and jumping into the open-source world brought lots of changes. It helped me realize my passion toward data management system. This was fantastic and fun experience and I will be continuing this journey.