28/03/2026 5:50 pm
Topic starter
Notifications
Clear all
Agent Skills Forum
3
Posts
4
Users
0
Reactions
71
Views
0
I am finally diving headfirst into AI after spending the last four months grinding through math and stats courses online. Im based in Chicago and Im really hoping to land a junior role or at least an internship in the next six months or so. I have been doing a ton of reading and obviously everyone says Python is the king because of libraries like PyTorch and Scikit-learn but then I run into these hardcore dev forums where people are arguing that Python is too slow for real production stuff and I should learn C++ instead if I want to do anything with computer vision.
I even saw some guy swearing by Julia for its speed with matrix math which just added more to my plate. Its kinda stressing me out because I dont want to waste my limited time learning the wrong thing if companies are looking for something specific. If Im trying to get into object detection and maybe some robotics down the line should I just stick with Python for now or is it actually necessary to be a pro at C++ too? I am just so pumped to start building but I dont want to hit a wall because I picked the easy language over the one that actually gets used in the industry. What do you guys think are the absolute must-haves for someone in my spot?
3 Answers
12
28/03/2026 5:50 pm
In my experience, Python is non-negotiable for landing that first role. Its the industry standard for research. But for robotics and vision in production, you'll eventually need C++ for that speed. I've tried many setups and this works:
- Python for prototyping with O'Reilly Hands-On Machine Learning with Scikit-Learn Keras and TensorFlow 3rd Edition
- C++ for deployment on NVIDIA Jetson Orin Nano 8GB Developer Kit Focus on Python first to get hired tho.
11
28/03/2026 6:05 pm
Python is the absolute backbone for a junior dev, especially for getting those libraries to play nice together! If you want to jump into vision without breaking the bank, I highly recommend picking up a Raspberry Pi 4 Model B 8GB RAM. Its fantastic for learning how to optimize Python code before you even touch C++. For the robotics side, the Elegoo Smart Robot Car Kit V4.0 is an amazing budget-friendly way to see your logic move a physical object. Dont get distracted by Julia right now... its cool for researchers but wont get you hired in Chicago as fast as Python will. Stick to basics and maybe grab a Logitech C920 HD Pro Webcam to start feeding real-time data into your models. You can honestly build a world-class portfolio with just these affordable tools!
2
28/03/2026 6:15 pm
Python is the right call for your first role, but you eventually need a mix:
- Python for high-level logic and research.
- C++ for deployment and robotics speed.
- SQL because you will spend half your time cleaning data. Dont stress over Julia yet. If you want to build, honestly any GPU from NVIDIA will work fine for local testing. It makes the CUDA learning curve way easier.
