Episode 3 is finally here! 🎙️🔥

I sat down with Mulu Tsega aka Luna for a raw and fun conversation about LinkedIn, finding jobs and opportunities, building a personal brand, and her journey from coding to product management.

Watch the full episode here 👇
🔗https://youtu.be/utJVxy2NnB0

If only I can put some people in .gitignore in real life😒

Okay, this is finally up.

Not a men vs women thing, not a lecture. A lot of us live in a bubble, tech, creative, liberal spaces, where things look fine for women, but there's a mindset that follows us out of it and into our own heads. I wrote a piece to show what it's actually like to move through the world as a woman, so you feel it instead of being argued at. Including what the "man-hater" is really reacting to, because it isn't hate, even when it sounds like it.

Please read the whole thing, it's one picture, and the pieces only add up together. Half of it out of context reads wrong.

If you've ever been confused why women say it's still not okay when it looks fine to you, this is the walk-through.

https://frostysiren.substack.com/p/on-being-seen?r=8ulkmq

And if it resonates, share it with anyone you think it might reach, women who'll feel seen, and men who genuinely want to understand

I just read this. And honestly I haven't read nor heard something that resonates so deeply as a woman. she has put something most women always felt but couldn't put in to words so clearly.
And please anyone here, read it.

Thank you❤️

I tried to quote one of my favorite section, but everything becomes my favorite

ማብሪያ ማጥፊያውን 100 times mokerku.I brought chair, climb like a monkey and tried to fix the bulb incase it was loose. And i was disappointed thinking i need new bulb. Then I tried to charge my PC, that's when I knew the light had been gone all this time. Moral of the story

እያበድኩ ነዉ

Death Of An Artist

By Alexander Raju

If death's inevitable, die at the right time,
While people retain you dear to their hearts;

Don't die too young, before your talents proved
That your death a devastation to the world;

Don't die too old that you a skeleton turned,
Your body shrunk to a shrimp, and face dried;

Die when your image is quite fresh and clear
In the minds of your kith and kins so dear;

Die when others could say, ‘early a bit'
That a few drops of tears would stain their cheeks;

Do your best and leave your life incomplete,
Give yourself a chance for a life second;

Hate the longevity of Tithonus,
Be Ulysses and conquer your lifespan;

Live and die a beauty, and be content,
For singing you stopped when your voice was good.

These past few days, I've been trying to understand how a machine learning model actually learns from its mistakes. And the easiest way to imagine for me was using a linear regression. So I tried to build linear regression with out scikit-learn library to learn what actually happens behind .fit().

And don't mind the dataset.It is intentionally small,one feature (house size) and one target (house price).

So normally, scikit-learn, training a linear regression model is as simple as:

model.fit(X, y)

Behind that single line, a lot happens: predictions are made and how wrong they are gets measured, the parameters figure out how they should change, they get updated, and the whole process repeats until the model finds a line that fits the data.And what I tried was to do these manually, just to see what happen.

So I implemented the prediction function (this was simple linear equation), Mean Squared Error (to see how the model prediction deviates from the real expected value), Gradient calculation and Gradient Descent (to determine how the parameters should change to reduce the error) and Parameter updates.

At the beginning, it knows absolutely nothing. It starts with a terrible line, measures how wrong it is, makes a tiny adjustment, measures again, and repeats that process hundreds of times. Learning is simply the accumulation of many small corrections.

To make that easier to understand, I also created an animation showing the regression line gradually fitting the data while the cost keeps decreasing.