4 ms·

> We're not quite at the level (yet) of feeding a whole codebase to an LLM and making it add features or make changes while understanding the big picture of the

> We're not quite at the level (yet) of feeding a whole codebase to an LLM and making it add features or make changes while understanding the big picture of the problem being solved We are one step away from that. All we need is a more advanced form of fine tuning.

r3d0c 3y ago
lol, this magical thinking; also not AI, it's ML
- culi 3y ago
  ML is a kind of AI
  - r3d0c 3y ago
    nope
    - culi 3y ago
      Sorry to do this to you, but you're actually incorrect.
      - r3d0c 3y ago
        sorry to dither your ego, but it's not; prove it, but i guess lying to yourself is easier than accepting you're wrong