A guide for people who want to self-study the basics of computer science

So I made a commitment: I decided I would work through Khan Academy math for 1-hour a day for 1 year. I started with pre-K [1] (specifically counting) and watched every video and did every single exercise in order. I focused on mastery. I didn't rush myself, and I did not continue until I felt completely confident in the material. I just did this for a year. I think I go through roughly algebra 2. In my mind, it is critical to combine explicit knowledge (watch videos) with tactic knowledge [2] (do exercises). For example, you need to understand what a logarithm is conceptually but you also just need to do problems to get a feel for it. So this is fundamentally different than learning-by-grazing or just reading a book.

I could go on and on, but let me just say that it changed my relationship to math in a deep way.

[1] https://www.khanacademy.org/math/k-8-grades [2] https://commoncog.com/tacit-knowledge-is-a-real-thing/

2) Build your similar own projects as well

3) Try with electronics. For that age group I have had good succes with MicroPython on ESP32’s + various sensors/actuators. The first thing I do is connect an LED strip with RGB LED’s and let them play with that.

4) Find him peers with the same interest. In that age peers are better partners. You can still help facilitate a bit, but the best would be to find a Coding club.

Part of it is that there's also the question of what your goal is.

If you down an MIT CS+EE degree curriculum in OCW, you're actually not going to have a huge amount of programming except as a byproduct. If your actual objective is to learn front-end development, I'm guessing you won't actually have learned that by the time you're done. I'm guessing most people looking for a self-directed CS bootcamp don't actually want a "CS" bootcamp.

Introduction to Mathematical Thinking by Keith Devlin (Stanford) https://www.coursera.org/learn/mathematical-thinking

You can probably retire on a 6% coupon if you save a lot, but you'll have a much more comfortable retirement if you can clip a 15% coupon on those same savings.

Discrete Mathematics is usually some mish-mash of formal logic, induction, combinatorics, number theory, and baby probability. You can get away without knowing these things, but studying them doesn't take very long (14 weeks, part time?!) and pays a very handsome dividend. A foundation in this sort of knowledge, and practice operationalizing it, is often the difference between $60K/yr and $100K+/yr for junior developers. The size of the dividend payment generally increases as one's career progresses, assuming you stay on the IC track.

Induction and formal logic are useful for reasoning about programs that contain loops. Often you can get away with intuitive understanding, and many people program for years before encountering problems where formal and structured reasoning becomes necessary (myself included!). But most well-paid developers will encounter a problem or two each year where reasoning about non-trivial loops is required and where the basics of induction and formal logic are tremendously helpful.

Exposure to basic thinking techniques in combinatorics is essential for reasoning even informally about time and resource complexity. Writing code that relies on non-trivial time and resource complexity analyses is occasionally necessary in most well-paying developer positions.

The very basics of number theory is useful for understanding, conceptually, how cryptography works. This isn't essential for most dev positions. But it is helpful knowledge.

Baby probability is increasingly required knowledge for many dev positions, and not just because of the explosion of AI/ML/DS.

Again, you don't need this material often. You can probably go an entire career avoiding the brutal pain of spending 14 weeks, 1-2 hours per work day, studying simple mathematics. But the ability to pull these skills out when they're needed is the difference between commodity "keep him around if we have billable hours" bootcamp labor and in-demand "smart and we really need to keep him around and eventually promote him to principal" labor.