# Found and lost

## Found

In my last post I lamented being a maths and stats graduate who realised he has no passion for data science, which is sort of what I've been gunning for for the past 4 years. I realised at the end of my second year that I do not like maths, realised after my placement year that statistical forcecasting is stupid for many applicaitons (thanks Taleb), then realised at the end of my final year that I do not like statistics at all. I like robots, so I had been forcing myself to believe that I liked neural networks and AI and what not, since it gave me a statistics based route into robotics, but quite frankly, I don't like AI either! I was utterly lost.

Then I was found. During my placement year I worked heavily on two projects which I thought of as robotics projects: a robot arm I designed and 3D printed, then programmed with an analytical inverse kinematics solver. If you don't know what that means, basically, I sat down and solved some trigonometric equations which figure out how the joints in a robot arm should bend in order to get the hand into a desired position and orientation. I then wrote a program to do this when given an orientation. I also tried to design a quadcopter flight computer. I succeeded in designing a PCB and writing the code to get the microcontroller on the PCB to read data from the accelerometer and gyroscope, then combine them into angle data using a Kalman filter... but the motors didn't work.

Anyway, what I realised was that I enjoyed these projects. What I realised was that when I was a kid I always wanted to be a programmer. What I realised was that I bought my first Arduino aged 16 and always wanted to learn more about electronics. What I realised was I've always wanted to read the Dragon Book, and learn how computers work. What I realised was that I didn't feel anything similar about statistics. I studied the wrong bloody degree!!

Hope, however, was back on the menu. I looked around for some embedded software engineering graduate roles and found that I better matched the person specification for those roles than I did for many data scientist ones! I decided I would learn all that I could about embedded development to make up for time lost, practise my C, and work on some projects. Hope!

## Lost

So how did that go? Well, I have learned a lot. I've been working on a 3D rasteriser that runs on an RP2040 connected to an LCD. I have learnt so much about C, so much about the Cortex-M0+, so much about low level computer operation, peripherals like SPI, things like DMA, and more! There's still a lot more to learn, but I think I can do it.

Why is this sectoin titled "Lost" then? Because I feel pathetic despite all this! I wake up in shame. I am an **idiot**. How could I study _the wrong degree_? Why would I do this to myself? Everything that I'm leraning right now I should've learnt during those four years I wasted learning what feels like nothing. Last year, I studied Decision Theory - what a waste of time! Uninteresting, useless subject - the only interesting parts are the epistomological bits, or the Jaynesian perspective - maybe I would've been better off studying bloody philosophy and reading Jaynes than doing a maths degree. I studied a whole Machine Learning course which taught me stuff about neural nets I already knew and a bunch of useless old techniques I couldn't care less about. What a waste!

You may feel I'm being too harsh on these subjects, and you'd be right. But this is _my_ blog post. I get to say how I feel, not what is necessarily true.

Anyway, I feel pathetic. I feel like I'm still never going to get a job. There's some jobs, sure, but I'm not that guy. I've realised a lot of my past projects, like that failed quadcopter one, are a bit crap. I also don't have any relevant internships under my belt, and the _wrong bloody degree_! I worry that I'll never get hired. I know this is stupid, but I worry anyway, and it makes me feel a great shame. I am a _failure_.