\
  The most prestigious law school admissions discussion board in the world.
BackRefresh Options Airdrop

I'm spending more time optimizing my harness than building anything

tp
180 fishy garrison puppy
  08/26/26
I mean that's how it should work
"in real life"
  08/26/26
...
180 fishy garrison puppy
  08/26/26
the only time an LLM should really be used to code anything ...
Cerise fortuitous meteor double fault
  08/26/26
...
puce exhilarant area corn cake
  08/26/26
rude
Crawly Twisted Dopamine
  08/26/26


Poast new message in this thread



Reply

Date: August 26th, 2026 12:10 PM
Author: 180 fishy garrison puppy

tp

(https://xoxohth.com/thread.php?thread_id=5897457&forum_id=2#50094217)



Reply

Date: August 26th, 2026 12:15 PM
Author: "in real life"

I mean that's how it should work

(https://xoxohth.com/thread.php?thread_id=5897457&forum_id=2#50094225)



Reply

Date: August 26th, 2026 12:17 PM
Author: 180 fishy garrison puppy



(https://xoxohth.com/thread.php?thread_id=5897457&forum_id=2#50094228)



Reply

Date: August 26th, 2026 4:27 PM
Author: Cerise fortuitous meteor double fault

the only time an LLM should really be used to code anything is for tool calling. 70+ on SWEbench is good enough IME

(https://xoxohth.com/thread.php?thread_id=5897457&forum_id=2#50094862)



Reply

Date: August 26th, 2026 4:12 PM
Author: puce exhilarant area corn cake



(https://xoxohth.com/thread.php?thread_id=5897457&forum_id=2#50094843)



Reply

Date: August 26th, 2026 4:23 PM
Author: Crawly Twisted Dopamine

rude

(https://xoxohth.com/thread.php?thread_id=5897457&forum_id=2#50094854)