Recently, I had a student job interview, for a position of a web developer. The company builds business applications, mostly for tourism sector.

It started of with a bit about me, my interests, previous experience, and knowledge. The requirements in the job posting included low code tools, so I thought I could be a good pick for the job. 

When it came to a question of if I know how to actually code, I said - mostly with AI assistance.

As it turns out, and which now makes sense, it is always expected that one understands the actual code. All of it.

The interesting part came towards the end. He said: "I don't know where to put you." Then, added: "I can see you know a little bit of everything, which I am not saying is good or bad, but I as a mentor don't know how I would mentor you".

I knew that he is right.

I did proceed to give him some more context. Considering that I am working on a start up, requires knowledge of a little bit of everything. As every start up is dynamic, and by using AI tools daily - I am kind of stuck in the chaos.

And it is true that without actual focus, without that single thing you are really really good at, like top 5-10%... You are nothing special.

Joker of all trades, master of none.

The whole situation reminded me of a thing I heard long time ago: let go some of your interests so a few may flourish.

But then questions about career focus start coming up.

Typical ones: Is this the right choice? Will I enjoy it? What if this won't be a prominent industry in the future?

The gut ones: Will I stick with it? How long will I stick with it? Or... is this just another "hobby"?

The more complex ones: How does this choice intertwine with my other choices? Will I advance in other areas, if I give my all into this one?

What makes it even harder is the scale of every field today. Especially when you are into something that's growing, like AI engineering. Even if you stick to a subtopic, like building great websites with AI... There is still infinite depth to it. Every element can be a specialization on it's own.

A good start is picking something specific enough, and throwing AI out of the equation. 

Lets say this is: "building the best websites for entrepreneurial events". Now why is this good and bad at the same time?

It's good because we have specified what we do.

It's bad because "entrepreneurial events" is still too broad. 

Under each entrepreneur there is a lot of complexity.

The goal is to have a group of people you can serve. It must be specific enough to be able to establish patterns from your past works. 

Let's take something as simple as a colour choice. You worked on a 100 websites, and you notice that blue tends to bring the highest number of conversions. 

Of course, there are thousands of other parameters, which would be added to an actual formula to measure the impact on conversion, however let's just ignore that for a second.

Now we have something to work with.

Your experience says that blue is the best choice. This means you have the first block to build on. Your principle states - pick blue; unless you are convinced, or have tested that you may reach better results with another colour.

But even with colour down, there is Pandora's box of questions, starting with: "What is a good website". 

The whole process of creation is... Complete chaos with a belief you'll find your way through. 

So, accept it's chaos. Build principles from your experience. Write them down. Stick to them, unless you discover a better way.

The best you can do, is to find a few blocks to stand on.