Kai Sassnowski and 3 others

Kai Sassnowski and 3 others

@sassnow.ski

I build stupid things with code. Writing interactive blog posts (veeeery slowly) at: https://sassnow.ski

1.8K followers144 following919 posts

Pinned

Alright, let me introduce myself on this platform. I'm Kai. I really like programming. Here's a list of projects I've done: A highly interactive blog post about 2D collision resolution in video games. I hope to finish part 2 this year, but no promises 🙈 www.sassnow.ski/rigid-body-c...

sassnow.skiWhat is a collision?From Mario bouncing off a Goomba to two cars bumping into each other in a racing game, dealing with collisions is such an integral part of most video games that we often take it for granted.

Kai Sassnowski and 3 others reposted

Kai Sassnowski and 3 others reposted

Replying to @freya.bsky.social

there used to be a strong correlation between seeing neat things online, and the person who made it: 1. is very knowledgeable and passionate about it 2. put love and care into it 3. is getting recognition for it but that trust is now gone, the grifters skipped to step 3, the slop ghouls have won

Kai Sassnowski and 3 others reposted

this wave of AI has legitimately killed a huge part of what I loved about seeing stuff online, what I loved about being an educator, and what I loved about sharing my own work I just don't feel like doing it anymore

What's making me really sad in the age of AI is that it completely removed this sense of wonder and awe from people's side projects for me. I really miss the "Holy shit, someone's brain actually did that" feeling I used to get. Like, now I see some really impressive projects and just go "eh".

Even if the result is 100% byte-identical (to use a Claude-ism), it's just not the same in a way I'm having a really hard time putting into words. I guess I'm finding that I was always way more fascinated by the process rather than the result.

On a more positive note: I had my first CS exam on Saturday ("Intro to Imperative Programming" lmao). It went... as expected. Exam was set for 120 min. I was done after 35. I got *really* paranoid for a bit that I had, like, missed another half of the exam or something 😅

I'm mentoring a lot of junior devs at my current job. By far the most common design advice I give them is to separate the thing that produces a value from the thing(s) that consumes the value. It's incredible how often that ends up massively simplifying the problem space.

I always love randomly seeing the things I’ve built in the wild 😊 (in this case utgars-chronicles.app)

The Unexplored Places@unexploredcast.bsky.social

the first of many upcoming Behind the Scenes posts for the new season is up on Patreon now! check it out for our whole spread of virtual notecards from our game of Microscope Chronicle in Appendix A: The Emerald Bloodletting www.patreon.com/posts/155699...

After reading a lot of mathematical proofs, I've come to the conclusion that every mathematician really wanted to be an English (or in my case, German) professor. Also, the subjunctive is now my best friend.

Kai Sassnowski and 3 others reposted

I spent 2 months learning about quantization and am extremely proud of the post I've written about it. I think these are some of the nicest visuals I've ever made, and I love how this compression technique invented in 1898 is being used on the bleeding edge in 2026. ngrok.com/blog/quantiz...

ngrok.comQuantization from the ground up | ngrok blogA complete guide to what quantization is, how it works, and how it's used to compress large language models

Even working through only the first few problem sets, I realized how much I love (and missed) chewing through difficult problems. There's something so satisfying about really having to think hard about a problem

It has begun! I'm especially looking forward to "Introduction to Imperative Programming" (I had to take CS as a minor) 🙃 At least they're using Go!

My biggest personal annoyance with AI is how no one seems to be building interesting things anymore that aren't AI-related. There's really only a handful of accounts left that I follow which feel like the pre-AI hacker space I used to love.

The problem with all these AI projects is that they're all so fundamentally .. uninteresting (to me). "My agent built an agent harness so that my agent can agent even harder (with AI!!)". Yeah, cool bro.

Part of the magic of people building stuff pre-AI for me was that no matter how silly the project, someone actually had to think about it and spend their own time building it. "Claude spit this out in 3 minutes" really doesn't hit the same.

Kai Sassnowski and 3 others reposted

I am happiest and proudest of my work when I am building things for other developers. I'd love to begin contributing to a meaningful, open source project, but I'm not sure where to even start.

If you would've told my 16 y/o self that at 35 I would get up at 4am so I can get in 2 hours of math studying before I head to work (and genuinely looking forward to that), I would've told you to lay off the cocaine for a bit.