CoCalc is likely the best online environment for project-based work, even for undergraduate students like myself. In our class, we have to train large machine learning models on Python notebooks, so we wanted to look for an easy-to-use online solution where we could collaborate and rent powerful compute servers for training. Not only was the setup and documentation easy to follow, the environment and tools that CoCalc provides are top-notch. You can tell that CoCalc was built with care and passion.

A few days in, our group faced a problem with compute servers (completely our own fault, I may add). We contacted support, initially expecting the usual long, drawn-out exchange with support staff to debug our issue. We never expected that the CEO and founder of CoCalc himself would give a practically instantaneous reply with his insights on our issue. In fact, he nailed our issue on the first try, and we have not had a problem since.

CoCalc has been an incredible resource and force multiplier for my undergraduate studies with its ease of use and powerful capabilities. There were some issues due to our relative inexperience with remote platforms and Linux servers, but I believe the amazing support provided by CoCalc wholly negates any concerns we have from now on. I would wholeheartedly recommend this to any student considering CoCalc as an extension of their local environment for studying and practicing machine learning.

I do different things with CoCalc. Right now, I am editing a shared LaTeX document which is a math paper.

I also use it for simple calculations with rational functions, which I would screw up if I did them by hand. Recently I used it as a convenient interface to Macauay2, which it is great for since I can never figure out how Macaulay2's interface is supposed to work.

All that to say, I feel like I am using only a tiny fraction of the power of CoCalc, but it is making my life easier, and I am grateful for it.