Learning is a key part of any career path, even more if you work in tech due to the massive amount of new technologies that arise every single month. When it comes to Machine Learning and AI, well, I don’t even need to repeat it after how exiciting this new year looks after ONE SINGLE MONTH 😱

But learning, specially with the fast peace of tech industries, can definitively feel overwhelming. This is why this is the first of three posts about learning that will be based on the: Why, What, How framework

Why should you learn?

Most people would answer to this with: because is good for your career but I’ve never found that answer to be very motivational plus is only “work realted”, and there is much more out there. In my opinion not learning is somehow to give up. There are two angles to this: Personal and profesional one.

Learning in a personal enviroment

One can argue that learning is an attitude to face life. If one looks to the scientific method, is all about learning. Make a hypotheis, create a experiment to validate this hypotesis, was the result expected? Yes -> great we know why, no, do we need to chanfe the the hypotheis or the experiment and keep in this loop. (MIRAR ESTO BIEN)

The reality is that you’ll need to face many unexpected unmangeble situation sin your lifetimeg and learning will be the key thing that will let you overcome those.

Learning in a profesional enviroment

Where does the compund effect happen?

If the intrisic reason to learn doesn’t convice you, maybe then is time you do it for something more tangible, like your own benefit. (ALI ABDAL)

There is a lot of literature around the power of habits (ATOMIC HABITS), but my preferred one so far has been (ZETTELCATESTEN). It falls a bit more on the How to learn section, but it ilustrated the compund effect very well.

Plus, it liustrates this with graph theory, which is always a plus for me. If you learn one new concept a day, and you connect it with three other ideas, in one year you’d have learn 365 new concepts and would have 1095 new connections. This numbers are big, yes, but let’s now think of a graph of 365 nodes, where each node is at least connected with 3 others. If every node has an average of 6 conections, simple paths (up to 10 nodes) will have 452 million possible paths.

** Note: This is using some aproximations such as:

avgDegree = (2 * edges) / nodes

(2 * 1095) / 365 ≈ 6 connections per node on average

possiblePaths = nodes * (avgDegree)^(L-1) / L
 L (from 2 to 10 nodes)

If that is not compounded enough for you, let’s look at another approach! Interests

Interests of sharing your knowledge

The price to pay here is sharing. but as the book (steal like an artist).

When you share your knowledeh and put out there in the internet you are leaving it to gain interest of work you’ve already done. A personal example is that some time ago I gave a talk on setting up and ML tool. Years later, while onboarding to a new role, a new collegue on day one told me: I’ve got the perfect project for you: we need that tool.

Now imagine that colleague was on the hiring pannel, he’d probabbly be biased towards hiring me, maybe other candidates where as good as me, maybe they were even better with that tool. But he could only find me.

Besides the personal anecdote the key thing is that an investment of time that happened 4 years ago counld have possitive impact in your carrer many years later.

Refrences

1. Simon Sinek’s book
   Start with why [Zeigarnik effect] -> learning is better when is interrupted