The Lean Startup of Skills Development
Given the pace at which technological developments are progressing and a never-ending amount of things to learn and un-learn, understand and implement, in-order for one to stay ahead of the competition in the technology job market, one can compare one's own progress in the path of learning and un-learning to a lean startup, which do not have much statistics or data on which feature or product will be a success with users, which will be a failure, what will fetch the most Return on Investment but the founders can only experiment by doing and un-doing things and analyzing the demand in the marketplace.
There is always a conflict between what the founders of a startup thinks would be a hit and what actually becomes a hit (the ecosystem do not need it, or it might be ahead of its time). Similarly an engineer with interests and skills in quantum computing might not be in demand in the current software job market or a few years back people with expertise in deep neural networks or robotics and computer vision were not so much in demand as compared to JAVA or .NET developers but now you know what is in demand. Both startup management and skills development involves a lot of uncertainty and dynamic market requirements. Things which are relevant today, might not be relevant tomorrow and if a startup does not adapt to changing ecosystem or a person do not develop the skills for tomorrow they are both going to die out. Google started as a search engine, now serving ads and building machine learning platforms. FB started as a social network, and now they are also serving ads, putting AI into photos and videos.
There is another angle to the above comparison, both startups and engineers are deriving the data about market requirements from what is trending in social media. I remember when suddenly everyone in India started getting into e-commerce. There was a selected few e-commerce companies in the beginning but suddenly there is large funding round to one of them and boom !!! every future seeker started to get into e-commerce because news about funding rounds spreads like wildfire on social media. Every engineer started to prepare themselves to clear their interview rounds even if the engineer had interests in machine learning but would eagerly work with web development PHP frameworks (thats me, by the way).
Suddenly a few companies in US decided to use latest open source technologies in the existing financial process and thus the term "FinTech" was born. Silicon Valley got funding, and again the future seekers from India started to disburse loans to individuals who are illegible to get loans from bank. Some of them got funding and some did not. But that did not stop individuals like me to explore bitcoins and the blockchain or credit rating algorithms because that is what was "hot" at that time. In came artificial intelligence and machine learning. Time for everyone's old academic interests to relive again. AI is here to stay but till how long nobody knows. While I was reading up about Haskell programming language (used by financial companies), the demand shifted towards Deep Neural Networks, CNN, RNN, LSTM etc. and thus Haskell and Java made way for R and Python.
"Change is the only constant"