Unless you’re training some ml models or doing heavy graphics, you might not need super heavy compute/gpu. If you do need, then remote into a desktop or vm is another option.
I have for the past 5 years or so focused on battery life over compute for my machines and then used RDP to access a VM dev machine hosted in azure. It’s worked great, and I was able to use free azure credits that came with msdn. You can do similar with ec2 and google compute. There are also container based dev environment services out there now. The benefit being you don’t need dev cruft on your everyday carry laptop. Plus if you get a new laptop or have multiple, your dev environment doesn’t change.
That said I did just switch to an m1 max as I also need to do Xcode. That has been a solid machine, though might be above budget. You could run Linux on it pretty sure.
I have for the past 5 years or so focused on battery life over compute for my machines and then used RDP to access a VM dev machine hosted in azure. It’s worked great, and I was able to use free azure credits that came with msdn. You can do similar with ec2 and google compute. There are also container based dev environment services out there now. The benefit being you don’t need dev cruft on your everyday carry laptop. Plus if you get a new laptop or have multiple, your dev environment doesn’t change.
That said I did just switch to an m1 max as I also need to do Xcode. That has been a solid machine, though might be above budget. You could run Linux on it pretty sure.