hey,
this week felt completely different to the last few, in a good way. i finally made some real, visible progress.
on mathhacks, i locked in the stack. i’m going with fastify (with typescript) for the backend so i can build a proper api that runs user code, checks it against test cases, and returns whether everything passed or failed. for the frontend, i’m sticking with nextjs + shadcn + tailwind, but i’m experimenting with supabase for the database and authentication.
the coolest part: i actually built a super simple code runner. i can now input a python function (like an addition function), run it against test cases, and instantly see which ones pass or fail. i even threw together a quick ui page to test it, and it works way better than i expected. learning how docker fits into all of this has been wild, and next week i’m planning to set up a proper waitlist for mathhacks and start adding multi-language support.
i’ve also started doing advent of code. it made me realise that my understanding of efficient code wasn’t as strong as i thought. those problems force me to think more about time and space complexity instead of just getting something working. it’s been uncomfortable in a good way - and i can already feel it making me a better programmer.
on the b2b idea, i haven’t built it yet, but i’ve set a clear goal: during the upcoming holidays, i want to ship a simple mvp and get it live. once it’s up, i’ll start trying to distribute it to real companies and see if i can turn it into my first real bit of revenue.
this week felt like a turning point. less dreaming, more building. less “one day”, more “this is actually happening”.
thanks for following along,
vulcan