I used to hate Python. Coming from C++ and Java, it felt fragile, inconsistent, and way too forgiving. Indentation defines scope, types are optional, performance isn’t great… and don’t get me started on packaging. The interesting part is: most of those things are still true. In today’s video, I talk about why I still use Python anyway, and the bigger lesson behind it. At some point, you realize it’s not about finding the “best” language. It’s about understanding trade-offs and choosing the right tool for the problem you’re solving. If you want to grow as a developer, that shift in thinking matters much more than the language you use. 👉 Watch here: https://lnkd.in/eJtP_jHF. #python #softwareengineering #programming #developers #careergrowth
python is for kids
Very similar experience here. 20+ years of experience with Java, solid background in other languages such as C++ and C#. When it comes to tools supporting DevOps processes, migrations, automation, and similar tasks, Python is my first choice. It enables compact, elegant code with minimal boilerplate. While Python’s runtime performance is significantly slower than native languages such as Go, this is often irrelevant for many of the use cases mentioned above. In practice, overall performance is frequently determined by factors such as network round trips, effective use of parallelism etc. I recall one case where AWS DMS failed to migrate a database table with tens of millions of records correctly (after 20+ hours, one of the columns in the target DB remained empty). I replaced the process with a Python script built on SQLAlchemy and concurrent.futures.ThreadPoolExecutor, and the table was migrated correctly in under 40 minutes.
As a C++ developer I felt the same until I started seriously playing with it and creating educational material with it. Now I love the idiomatic constructs it introduces, and NumPy is a well designed, expressive API for applied math. And also, the way it makes autodiff possible, the way JAX works is very elegant.
People who hate Python are stuck in the late 90s. They still dont comprehend the main goal of programming languages, which is solving problems, it doesnt matter what language you use, as long as it achieves the objective flawlessly.
Python has democratized computer science, enabling scientists and builders to go from idea to prototype in hours, not months. It lowered the barrier to entry so much that almost anyone can become productive with minimal effort. And that’s exactly the point. With the Python ecosystem, complexity stays where it matters most: solving the problem. For the vast majority of applications, the inefficiency of the compute stack is irrelevant. Hardware is cheap, time is not. Optimization is only valuable once the problem is worth solving.
Every year I have interns during the summer coming from colleges and we pair program together for the entire two months. They always ask me, which is the best language? I always tell them it depends on what are you trying to accomplish. Every language has strengths, weaknesses and trade-offs. You wouldn't want to make a website using assembly language. You wouldn't want to program a hardware using HTML language. Think about what you want to build, how fast you want to build it and how it will scale (if it needs to scale). For an example, Python is good for automation, AI, and a lot of other things. However, what you're trying to accomplish, Python doesn't do a good job on, then you have to move on to a language that does.
Arjan Egges since Claude Code came into the picture, programing language has become irrelevant. I code in a "fragile" framework of C++ yet happily building stuff in Java,Kotlin ( for Android ) which I have not used in ages and converting them to Metal and Vulkan for macOs ( which I dont use anyway but which is needed for cross platform compatibility). This is the power of #agentic bots in coding.
yes.. I went through something similar. Took me 4-6m of intensive use. I now do not use anything else.
So relatable. In my circles, I am the biggest Java advocate there is — 20+ years of enterprise systems will do that to you. But the trade-off thinking you're describing is exactly what separates seniors from juniors. I've shipped Python for automation work where spinning up a Spring context would've been absurd. Right tool, right job.
Those who use Python are developers? ✅