I started my path as BE 5 years ago and now I run a blockchain company.
My strategy was:
1 - learn just the basics on each vertical (bitcoin, EVM, smart contracts, consensus protocols, game theory) - it wasn't difficult having a background as engineer
2 - get a job as PM/PO to be in touch with as many projects as I can
3 - get a deep understanding of the market needs
4 - start working as low level developer on a particular (selected) project, to understand the core and become an expert
5 - manage my own team to sell services (outsourcing) to other entrepreneurs
6 - scale num.5
First at all, a huge background on maths and statistics. Then, select a programming language and become as good as posible. I recommend you to choose python because it has a lot of libraries of data science (I recommend you to learn numpy, pandas, scipy and sklearn). After all, you should consider to put tensorflow in your learning curriculum.
DEX could be a good idea. However, I've tried something without success. In waves, for example, you can sell btc for USD tokens with good profit, but you can not extract them. When exchanging with other ctyptos there is no high volume, then I've found bid / ask prices with high spread.
That's the point, you can't have so many balances in so many exchanges, because, in that case, each return is going to be very small. Just to take profit on an ETH USD arbitrage, you have to have 25% of your capital in ETH and 25% USD in exchange A, and 25% ETH and 25% USD in exchange B. So, when you take advantage of the opportunity to buy ETH at low price on exchange A and to sell on exchange B, you are only earning an arbitrage profit on a 25% of your capital.
And that profit become less and less if you divide your capital into more coins and more exchanges.
My strategy was: 1 - learn just the basics on each vertical (bitcoin, EVM, smart contracts, consensus protocols, game theory) - it wasn't difficult having a background as engineer 2 - get a job as PM/PO to be in touch with as many projects as I can 3 - get a deep understanding of the market needs 4 - start working as low level developer on a particular (selected) project, to understand the core and become an expert 5 - manage my own team to sell services (outsourcing) to other entrepreneurs 6 - scale num.5
I share with you a doc I wrote a time ago, describing the roadmap I propose to learn the basics: https://medium.com/rather-labs/the-roadmap-to-become-a-block...
Also, if you want to check my company, you can do it at ratherlabs.com
I wish you the best on this exciting path!