Been working on my NumDuel project lately.
Started as a simple number guessing game in Python, but instead of jumping straight into FastAPI and databases, I'm trying to learn things properly and evolve it step by step.
Been working on my NumDuel project lately.
Started as a simple number guessing game in Python, but instead of jumping straight into FastAPI and databases, I'm trying to learn things properly and evolve it step by step.
this is how hints . py (space is intentional) looks and after half the attempts are exhausted by the user it returns the hint to main . py so that it shows the hint to the player whether the secret no is even or odd and other hint
#numduel
confest .py and test_scoring. py
while creating test folder learned about:
pytest is a testing framework. Instead of manually running my game and checking if scoring works, I write functions that check it automatically.
Today i learned about Brook's law According to Brooks himself, the law is an (outrageous oversimplification) basically it states this -->Adding manpower to a late software project makes it later
new convention according to me instead of using inverted commas i use brackets
I just read a research paper that might be the most quietly important thing in AI safety right now.
It's from Google DeepMind. And it says something shockingly simple:
Teaching AI to copy humans might be safer than teaching it to win.
Here's why that changes everything 🧵
Claude isn't being taught more chemistry. It's being taught to work with the same kinds of evidence chemists use.
A lot of chemistry is actually a detective problem: you start with clues and try to figure out what molecule produced them.