The branded technologies change with the breeze. This is not progress, or an emergent property of progress so much as it is by design (but that is a completely

The branded technologies change with the breeze. This is not progress, or an emergent property of progress so much as it is by design (but that is a completely different conversation). The best investment of your time is in the fundamentals; the things that don't change or change very slowly: Stats: Useful in both data science and machine learning Algorithms: Duh Differential Geometry: must-have if you're a graphics/games guy Discrete Math: Good stuff for dealing with machines that process things in bits. Linear Algebra: Comes in handy in so many different problem domains. Ethernet + TCP/IP: It's not going anywhere. Hardware/Architecture: Still running a von Neumann architecture last I checked. Things go much more smoothly when you know what the physical limitations and bottlenecks are before you start coding. If you go all-in on WebGL and get a job, keep an eye out for the next new hotness, because you will officially be on the treadmill then.