i came up with a 53-minute deep dive on the mini container runtime i built in c++. it's called dabba (डब्बा). i talk about namespaces, cgroups, mounting, networking (veth, nat, masquerading), and much more.
the post builds layer by layer and gives you an intuition for what a container is rather than just a definition. you'll understand a lot more than just the docker run command.
link to the blog post in the comments :D enjoy!
my bro has been cooking lately. i got this very cool card. (why is it commonnn ????)
4 days. Solo. $0 budget.
I built an app that turns what you love into a holographic taste card + the fashion brands your taste pulls toward.
The AI isn't allowed to pick a brand on its own. If the data didn't return it, it doesn't show up.
1
9
479
this is actually super nice dude
built SaaT (saat, turkish for time)
implemented 11 distributed clock algorithms from scratch in Go. zero frameworks.
from physical wallclocks to lamport clocks and vector clocks, covering google's truetime, interval tree clocks, and HLC. compared tradeoffs and how they resolve conflicts
also built a network simulator with jitter to audit causality and test concurrent edge cases across them
[ link : github.com/dexisback/Saat ]
1
21
1,106
reposting it for myself to read later.
pointers in c, hehehe
cintu07.github.io/posts/poin…
high time you check design index by my guy
Replying to @shadcnlabs
@shadcnlabs is now sponsoring designindex.xyz
Big fan of what Shadcn Labs is doing. They’re pushing the shadcn/ui ecosystem way beyond buttons and forms, with open-source components for AI agents, terminal apps, video, OG images, emails, PDFs, and MCP apps.
1
19
719
look at this, like honestly. it's an impressive collection of design resources and the ui is absolutely gorgeous.
putting a lot of effort into improving the search on designindex.xyz
2
18
1,599
the thing is, you can get a huge payoff from learning cs fundamentals close to the machine. ugh, some of this stuff involves actually studying how filesystems work, optimising log ingestion and data cleaning shit like that. i also remember there's a whole repo with inference implemented in a single file.
optimising shit, learning cuda maybe, studying how tiling and caching work. maybe that'll give me some intuition.
ai can't write beautiful software, but we can't limit our thinking to that. reading, observing, and maybe trying to make it sit there and come up with tweaks to make things better is what will keep us in this "race".
give it an hour or two every weekend, maybe?
adding a bit :
- hobbies are what you like, not what you are good at. honestly, a hobby doesnt need to be tuning your skill. it can be something arbitrary like noting down the patterns of bird activity outside your window.
- writing helps to think often, and if you write more, you think straight. same is the effect of talking to someone you are easy with and you trust with your vulnerability
2
37
842
















