Even without software dev.
Pattern seems consistent:
Identify ways to improve current product with AI. (Sprinkles)
Rethink what the whole thing is with AI and realize the original product kinda didn’t make sense.
Similar to how folks are analogizing AI to roles, and not skills, right now.
Once you really rethink a product in an AI world, it’s likely such a different product that you laugh seeing current adoption (like your vibe of self driving being like a virtual person. Why? Only because it’s our most basic analogy to change the current mode).
The current approach to building AI agents that write software is like building a self-driving car by putting a robot in the driver’s seat and having it physically turn the steering wheel and push the pedals
now raising capital for a babysitting startup where you get paid to have your baby taken to Europe for a week at a time
daily pictures of your baby in various landmark locations!
pro tip for Americans visiting Paris is travel with a baby. take a stroller & baby carrier.
museums let you skip lines. people on metro ask if you want their seat. complimentary wine. free pastries.
as a bonus babies are awesome. if you don’t have a baby, make / borrow one idc.
In Postgres `where column like ‘foo%’` efficiently gets optimized into a range scan over an index on `column`. `where column >= ‘foo’ and column < ‘fon’`
If GitHub Copilot is trained on existing data, why would it suggest the use of the literal value for e rather than the provided constant? Surely none of the code it's trained on does this?
Did you know that the US Govt has 24/7 high resolution video from any location on earth? They can select any date, time and location and see in fine detail what happened, even at night. It’s as if Google earth would allow you to jump to any point in time.
The story of Nigel Richards, the man from New Zealand who memorized every French word in the French scrabble dictionary and won the French Scrabble Championship without speaking any French
[read more: buff.ly/2rK5wi4]